The transition made by professional athletes from a full-time elite athletic career to retirement has received considerable comment in the sports media (4,5) but little formal study. It is generally believed that the transition is difficult and traumatic because of the sudden cessation of the intense demands of elite athletic performance, compounded by the sudden loss of the athlete's intense devotion to professional athletic competition and its attendant rewards. At the elite level, the athlete's life is fully organized around his or her performance, as are support personnel and extensive logistic and financial resources. On retirement, athletes have reported jarring transitions to a life in which the focus of such intense commitment is unclear, the resources and personnel that organized and managed their lives away from the competition venue are lost, and the rewards, both emotional and financial, are diminished. Previous studies have assessed the risk of suicide in active college athletes (10), the risk of depression in college graduates who were previously varsity athletes (13), the association between concussion experienced during active professional football participation and the diagnosis of mild cognitive impairment later in life (8), and analyses of case series of suicide in professional athletes (4) or injured athletes (20), but we know of no studies in which the prevalence and impact of depression and pain have been assessed in a more structured way among retired National Football League (NFL) players.
We surveyed retired NFL players who are members of the National Football League Players Association (NFLPA), regarding the life domains in which problems were encountered in making the transition from active competition to retirement, the magnitude and impact of such difficulty, barriers to seeking and receiving assistance in resolving these difficulties, symptoms of and experiences with depression and chronic pain, and recommendations for how these problems could be remedied and how retirement could be made more successful and satisfying.
The sample of subjects to be surveyed was obtained from the active membership list of the NFLPA, retired players section (NFLPA-RP). A total of 3377 surveys were sent to retired players via surface mail. Reminder postcards were mailed 1 wk later, followed by a second survey mailed to all nonrespondents approximately 1 month after the initial mailing. Follow-up surveys of nonrespondents to assess the comparability of respondents and nonrespondents could not be conducted, because of financial and logistic constraints.
Each survey contained a cover letter that described the purpose of the survey and assured respondents that responses were completely confidential and would in no way affect their membership in the NFLPA. Names were connected with responses in a master list of names and identification codes kept in a locked and secured location. The cover letter carried the NFLPA-RP logo and was signed by the executive director of the NFLPA-RP. A waiver of the documentation of informed consent was approved by the institutional review board of the University of Michigan (#2005-279), based on minimal risk and the documentation of appropriate procedures to maintain confidentiality of all responses.
The survey assessed the experience of respondents with a range of life problems following retirement, such as employment, marital or financial problems, the barriers to receiving help for these problems, and the types of programs that might be helpful for retired players. The survey included a structured depression questionnaire (see below), a single question regarding chronic pain, and demographic questions.
Depression symptoms were measured by the PHQ-9, a validated screening questionnaire based on standardized diagnostic criteria (11,21), including an assessment of the impact of depressive symptoms on personal and work roles. A self-rating of health status was made on a five-point scale, followed by assessments of past or current difficulties with nutrition, exercise, alcohol use, smoking, and depression. Respondents were asked additional questions about problems with alcohol and the impact of chronic pain on normal work.
The survey was pilot tested with the Detroit chapter of the NFLPA-RP, resulting in several modifications to its final form.
Simple frequencies and summary statistics were calculated on all variables. The PHQ-9 responses were calculated to create a binomial depression severity classification using standardized cutoffs (11,21) to distinguish between no or mild depression versus moderate to severe depression. A binomial variable was also created from responses to the item "difficulty with pain," with "very" or "quite common" considered high pain, and "somewhat" or "not common" considered low pain. Using t-tests, chi-square where appropriate, comparisons were made between all items and both the depression severity classification and the pain ratings. In addition, a similar analysis was conducted using a variable created by combining the two categorizations of depression (high vs low) and two categorizations of pain (high vs low), resulting in four mutually exclusive groups (high depression/high pain, high depression/low pain, low depression/high pain, low depression/low pain).
Usable responses were received from 1617 members (crude response rate, 47.9%), with a functional response rate of 48.6% when accounting for surveys returned undelivered (N = 36) or unable to be completed because of death (N = 10) or mental incompetence (N = 16). The mean age of all respondents was 53.4 yr (± 14.5), and 80%were married. Roughly 30% (N = 483) of respondents reported current involvement in football, most commonly through coaching at the high school or college level. The mean number of years for which respondents played professional football was 7.1 (± 3.6), and they had played for a mean of 2.3 teams (± 1.3) per respondent. The median time since retirement was 25 yr. Roughly a third each of respondents reported having been "cut" (N = 557) (meaning that they had ended their career not because of injury and not of their own choosing), "retired of my own choice" (N = 559), or retired because of a career-ending injury (N = 470).
Common Retirement Problems
The most common retirement problems reported by respondents were (in descending order of frequency as quite or very common): difficulty with pain (48%), loss of fitness and lack of exercise (29%), weight gain (28%), trouble sleeping (28%), difficulty with aging (27%), and trouble with transition to life after professional football (27%). The most commonly reported barriers to seeking help for these problems (reported as important or very important) were a preference to use spiritual means to deal with these issues (36%), preference to deal with these issues with family and friends (33%), lack of insurance coverage (33%), and lack of recognition that these problems were important (33%).
The mean PHQ-9 score for all respondents was 4.5 (out of a maximum total score of 27 ± 5.3). The proportion of respondents responding in the no-to-mild category of depression (PHQ-9 score 0-9) was 84.5% (N = 1366) and, in the moderate-to-severe category (PHQ-9 score 10-27), 14.7% (N = 237; there was no response by 14 respondents). Roughly 7% (N = 117) rated the impact of depressive symptoms as making work or home life very or extremely difficult.
The odds ratios for respondents reporting various transition and retirement problems if they reported moderate to severe depression versus no to mild depression are shown in Table 1. For example, respondents scoring as moderately to severely depressed were 11.2 times more likely to report trouble sleeping than those rated as not or mildly depressed, 7.8 times more likely to report a loss of fitness and lack of exercise, and 7.1 times more likely to report financial difficulties. Other problems reported significantly more commonly in respondents rated as moderately to severely depressed were lack of social support or friendships, the use of prescribed medication, alcohol, or other drugs, and trouble with the transition to life after professional football.
Similar to the analysis above, odds ratios for respondents reporting various barriers to seeking help if they reported moderate to severe depression versus no to mild depression are shown in Table 2. Respondents reporting moderate to severe depression were most likely to report that "I feel I would be weak if I got help," "I would be embarrassed by what friends or family would think," "help is too far away," and "I didn't recognize issues as important" compared with those with no to mild depression.
Difficulty with pain was reported as very common by 404 respondents (25.2%), quite common by 365 respondents (22.7%), and not or somewhat common by 837 respondents (52.1%). The odds ratios for respondents reporting various transition and retirement problems are shown in Table 3, comparing those responding that difficulty with pain is quite common or very common versus those reporting it as not common or somewhat common. The most common transition problems for respondents for whom pain was common versus those for whom pain was uncommon were difficulty with aging; the use of prescribed medication, alcohol, or other drugs; trouble sleeping; and loss of fitness and lack of exercise. The survey did not allow further analysis regarding specific types of injuries or sources of pain.
Interaction of Pain and Depression
Because of the common cooccurrence of pain with depression in the general population (2,6,12), the life experiences of respondents who had quite or very common difficulty with pain as well as moderate to severe depression scores (high depression/high pain) were compared with those having low scores in both pain and depression (Table 4). Of the total of 1594 respondents, 173 (10.9%) had high scores in both areas, 593 (37.2%) had high pain scores and low depression scores, 63 (4.0%) had low pain scores and high depression scores, and 765 (48%) had low scores in both. High depression/high pain respondents were 32 times more likely to report trouble sleeping compared with those with low depression/low pain, with high odds ratios for difficulty with aging; loss of fitness and lack of exercise; the use of prescribed medication, alcohol, or other drugs; and financial difficulties.
The comparable analysis for the barriers experienced by respondents, according to having either both high or both low scores in pain and depression, are shown in Table 5. The barriers reported as most common by those with high depression/high pain scores compared with those with low depression/low pain were "I feel I would be weak if I got help," "help is too far away," and "I didn't recognize these issues as important; thought they were a part of life".
Programs to help with the following problems were most commonly requested by respondents (percentage rated as very or quite helpful): programs to help with fitness and exercise (48%), nutrition (46%), financial assistance (46%), pain management (43%), relaxation (42%), distress or depression, (42%), and spirituality (41%).
These self-report survey data from a large sample of retired professional football players give a structured assessment of depression and pain symptoms experienced by professional athletes after retirement. Their problems are not necessarily worse than those experienced by the general population in retirement, nor are they necessarily better (9,15). Although it is difficult to make comparisons of this population versus other retired populations because of the difference in age at retirement, most studies would suggest that older adults have roughly similar levels of depressive symptoms, in the range of 8-10% as measured by self-report questionnaires (14,15). The high profile of many of these retired professional athletes, and the concomitant financial benefits and emotional support they experience, seems to neither increase nor decrease the likelihood that an athlete will experience significant difficulties in retirement, although we could not make direct correlations with income or retirement assets. The prevalence of moderate to severe depressive symptoms as measured by the PHQ-9 is also roughly similar to that found in the general population (17), and seems to be higher than in younger, active athletic populations (10,16). The association of certain problems (e.g., trouble sleeping, use of prescription medications, drugs, or alcohol, and loss of fitness) with higher depression symptom scores is also typical for patients in the general population, as are associations with a wide range of comorbid medical conditions, including diabetes, cardiovascular disease, cancer, and several neurological diseases (7).
It is the cooccurrence of depressive symptoms and pain that puts retired players at the highest risk of significant difficulties in retirement. The relationship between depression and pain is important and complex. A recent systematic review of almost 60 studies of the comorbidity of pain and depression showed that roughly two thirds of patients with major depressive disorder had significant pain symptoms, and roughly half of patients seen in chronic pain clinics met criteria for major depressive disorder-proportions that are roughly similar to the cooccurrence found in these subjects (2). For example, of 236 subjects in this study with high depression scores, 173 (73%) had high pain scores. Most patients eventually diagnosed as depressed in primary care settings present initially with somatic symptoms-most commonly, pain complaints, including back pain, chronic abdominal or pelvic pain, or headache (19). The biological and psychological mechanisms underlying these highly comorbid conditions have been studied in depth (12) and suggest that depression more likely follows pain than vice versa, and that the risk of developing depression is correlated with the severity of the pain complaints (6). Recent research has focused on approaches to screening and diagnosis of comorbid pain and depression (because there is considerable symptom overlap) and combined approaches to treatment using both medications and psychotherapeutic interventions such as cognitive behavioral therapy (6). It also seems that the baseline level of pain in patients with both disorders detracts from the eventual response to treatments for depression (3). These data can be linked with findings in a recent study in which vigorous physical activity was highly protective of the development of depression in older former athletes (1) to suggest that a hypothesis worth further exploration is that the high level of physical disability and chronic pain with which these athletes leave their football career causes them to have significant difficulty maintaining their activity and fitness levels, thus predisposing them to an increased risk of depression (14).
Although pain and depression are commonly comorbid in the general population (2,12), the frequency with which retired professional football players report difficulty with pain seems to put them at additional risk of both developing depression and experiencing associated difficulties with retirement. The high level of psychosocial dysfunction and significant barriers to receiving help put a small but important subgroup of all retired NFL players at significant risk of adverse life events and disability, almost certainly including an increased risk of suicide (4). The respondents with high scores for either depression or pain endorsed a congruent set of programs for life assistance, including programs focusing on help with fitness and exercise, nutrition, financial assistance, and sleep, marital, and relationship problems. Given the significant barriers to effective treatment noted above, and the known difficulty in treating patients comorbid for both depression and pain (6), future research should evaluate a range of clinical and educational outreach programs to serve this population, including ways to provide anticipatory guidance to active players as they plan for their future retirement. This is particularly true because retirement can occur suddenly, with little opportunity for planning, after a career-ending injury or being unexpectedly cut from the team. Possible approaches that deserve development and evaluation include educational interventions to make players aware of potential future problems, self-assessment instruments, educational resources, and a network of clinical services organized around NFLPA chapters, most of which are based in current NFL team cities.
The most obvious limitation to this study is the self-report nature of both the depression and pain data, as well as the respondents' perceptions of the life problems they are experiencing, barriers to receiving help, and the types of programs most needed. The PHQ-9 has fair concordance with longer diagnostic interviews conducted by trained health care professionals, but it is overly sensitive, with a significant rate of false positives based on self-report symptoms that may not reflect a deeper, more enduring clinical depression (18,22). Most studies in the general population suggest that roughly 25-40% of those screening positive for depression on any of several self-report instruments will have a criterion-based diagnosis based on a structured clinical interview (22). The single question asking about difficulty with pain is not likely to be as accurate as more detailed visual analog scales and other standardized questionnaires assessing pain, but it was adequate for our initial purposes in understanding the basic issues faced by retired professional football players. More detailed assessments and interviews on a smaller population, with more objective data concerning clinical diagnoses, financial status, or health insurance, for example, would be appropriate to characterize more fully the clinical and demographic characteristics of this study population.
The response rate of nearly 50% can be seen as literally either "half full" or "half empty" with regard to the generalizability of these results to the larger population of retired professional football players. Our inability to follow up with nonrespondents regarding their comparability with respondents is a significant weakness. Retired players who chose to not respond could be more depressed and have more life problems, making them less likely to take the time or have the energy to complete the survey. Or, they may be less depressed and have fewer life difficulties, leading them to believe that the survey does not apply to their life situation. A future study should explore methods to increase the response rate and make particular outreach to those who do not initially respond.
Retired professional football players experience depressive symptoms at a rate that is similar to that found in the general population, presumably with a corresponding rate of clinical depression. They bear an additional burden of substantial chronic pain. Depressive symptoms and pain interact to result in a strong correlation with self-report perceptions of the risk of sleeping problems, difficulty with aging, loss of fitness and lack of exercise, financial problems, and concerns about their use of prescription and recreational drugs and alcohol. Retired professional football players bear the same stigma as does the general population with regard to barriers to seeking helping for their difficulties with depression and pain, including not recognizing the issues as important, not knowing where to seek help, feeling they would be weak if they sought help, and being embarrassed by what family members and friends would think. One hypothesis worth further exploration is that the high level of physical disability and chronic pain with which these athletes leave their football career causes them to have significant difficulty maintaining their activity and fitness levels, thus predisposing them to an increased risk of depression. This population of retired professional athletes would benefit from a proactive educational and clinical outreach program, possibly beginning even before retirement on an anticipatory and preventive basis, to improve the likelihood that retired NFL players will achieve a high quality of life after football.
The authors acknowledge with gratitude the guidance provided by Sheila M. Marcus, MD, Kristen L. Barry, PhD, John F. Greden, MD, and Thomas Carli, MD (deceased) of the Department of Psychiatry and the Depression Center at the University of Michigan. Additional financial support was provided by the Depression Center and the Department of Family Medicine at the University of Michigan. The authors also acknowledge the considerable feedback provided by the Detroit chapter of the NFLPA, and the technical support provided by Alexis Ghee and Michael Rubyan.
This study was supported in part by a grant from the National Football League Players Association.
The authors report no conflicts of interest.
1. Backmand, H., J. Kaprio, U. Kujala, et al. Influence of physical activity on depression and anxiety of former elite athletes. Int. J. Sports Med.
2. Bair, M. J., R. L. Robinson, W. Katon, et al. Depression and pain comorbidity: a literature review. Arch. Intern. Med.
3. Bair, M. J., R. L. Robinson, G. J. Eckert, et al. Impact of pain on depression treatment response in primary care. Psychosom. Med.
4. Baum, A. L. Suicide in athletes: a review and commentary. Clin. Sports Med.
5. Burton, R. W. Mental illness in athletes. In: Sport Psychiatry
, D. Begel and R. W. Burton (Eds.). New York, NY: W. W. Norton and Co, pp. 61-81, 2000.
6. Campbell, L. C., D. J. Clauw, and F. J. Keefe. Persistent pain and depression: a biopsychosocial perspective. Biol. Psychiatry
7. Evans, D. L., and D. S. Charney. Mood disorders and medical illness: a major public health problem. Biol. Psychiatry
54: 177-180, 2003.
8. Guskiewicz, K. M., S. W. Marshall, J. Bailes, et al. Association between recurrent concussion and late-life cognitive impairment in retired professional football players. Neurosurgery
9. Kessler, R. C., K. A. McGonagle, S. Zhao, et al. Life-time and 12-month prevalence of DSM-III-R psychiatric disorders in the United States. Results from the National Comorbidity Survey. Arch. Gen. Psychiatry
10. Kokotailo, P. K., B. C. Henry, R. E. Koscik, et al. Substance use and other health risk behaviors in collegiate athletes. Clin. J. Sport Med.
11. Kroenke, K., R. L. Spitzer, and J. B. W. Williams. The PHQ-9. Validity of a brief depression severity measure. J. Gen. Intern. Med.
12. Ong, K. S., and S. B. Keng. The biological, social, and psychological relationship between depression and chronic pain. Cranio
13. Paffenberger, R. S., Jr, I. M. Lee, and R. Leung. Physical activity and personal characteristics associated with depression and suicide in American college men. Acta Psychiatr. Scand. Suppl.
14. Paluska, S. A., and T. L. Schwenk. Physical activity and mental health. Sports Med.
15. Pennix, B. W., J. M. Guralnik, L. Ferucci, et al. Depressive symptoms and physical decline in community-dwelling older persons. JAMA
16. Puffer, J. C., and J. M. McShane. Depression and chronic fatigue in the college student athlete. Clin. Sports Med.
17. Rief, W., A. Nanke, A. Klaiberg, and E. Braehler. Base rates for panic and depression according to the brief patient health questionnaire, a population-based study. J. Affect. Disord.
18. Schwenk, T. L. Screening for depression in primary care: a disease in search of a test. J. Gen. Intern. Med.
19. Schwenk, T. L. Diagnosis of late life depression. The view from primary care. Biol. Psychiatry
20. Smith, A. M., and E. K. Milliner. Injured athletes and the risk of suicide. J. Athl. Train.
21. Spitzer, R. L., K. Kroenke, and J. B. Williams. Patient health questionnaire primary care study group. Validation and utility of a self-report version of PRIME-MD: the PHQ primary care study. JAMA
22. Valenstein, M., S. Vigtan, J. F. Zeber, et al. The cost-utility of screening for depression in primary care. Ann. Intern. Med.
134: 345-360, 2001.