S O U TH A F R IC A N JO U R N A L O F PORTS MEDICINE Vol 4 No 4 N o vem b er 1997 Mcjgnesft* The first solution for magnesium deficienciesRe pr od uc ed b y Sa bi ne t G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) t o fiSSr»K L the edge improve bodyWork t j r v w v - • > * > M W J S t x & j > The multivitamin with extra magnesium that helps the body work. I- ' Madaus Pharmaceuticals (Pty) Ltd Reg No.. 77/00174/07 Cnr. 16th Road & 2nd Stieoi Halfway House 1685 G N E -A -D A Y M ultivitamin and Mineral Supplement ~ J 0 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) EDITORIAL Sport Psychology - Special Edition It is an cxcittng propsect that this edition o f SAJSM is entirely devoted to sports psychology. It marks another m ilestone in sport psychology's developm ent in our country. Since South A frica ’s re-entry to international sport, there has been growing interest in the d iscipline by athletes, coa ch es, sports adm inistrators and the m edia alike. Judging by the enthusiasm shown by students, there is a sim ilar k eenness to meet the consequent dem ands. W hile this b o d e s well for the future o f South African sport psycholog)', there arc many organisational issues within the d iscipline w hich still need to be a dd ressed. One thing is clear however: A m id st all the academ ic and professional deba tes there is no doubt that the b urgeon ­ ing interest has d evelop ed in response to a need - the need to understand and control the diverse influences which p sychological factors have on sports perform ance and the general w ell-being o f athletes. The varied nature o f this need is reflected in the diverse topics and styles evident in this edition. Pat Scott provid es a theoretical overview o f how p h ysiological and p sychological factors interact to make pain a com plex and often m isunderstood phenom enon. A nderson, Bassson, G eils and Farman’s article on personality style, m ood states and negative a dd iction , is representative o f a b ody o f research w hich has been included under botli sport and health psychology. It focu ses on the w ell-being o f the general population rather than sim ply the p er­ formance o f elite athletes. In contrast to the theoretical and em pirical based em phasis o f the previous articles, Clive B asson ’s case-study p rovides readers with an insight into applied sport psychology. It is a source o f practical ideas for the practition­ er and athlete. Finally, O livier looks at p hysiological, kinem atic and p sychological d ifferen ces betw een overground and treadm ill running. W hile psycholog)' is not the main em ph asis, the paper clearly d em onstrates the relevance o f sport p sycholog)' to the other sports s cien ces - viz. adding a new dim ension ( in this case the influence o f cognitive appraisal ) to our undertsanding o f con cepts, lessen s the variance w hich is unaccounted for in scien tific research. In light o f the above, sport p svch olog)’ expands sports s cie n ce ’s know ledge of, and ability' to contribute towards the w ell-being and perform ance o f atld etes. Should the traditional term “ sh rin k ” then not rather be replaced with “ expander” ... ?! Clinton Gahwiler Guest Editor Rep̂ nicGel On-the-spot pain relief. [sT| Reparil®-Gel 100 g contains: Aescin 1.0 g; Diethylamine salicylate 5.0 g. Reg. No. G/13.9/2367. Byk Madaus (Pty) Ltd. Reg. No. 82/11215/07. 2nd Road (enr 16th Road). Randjespark. Midrand 1685. SPORTS MEDICINE NOVEMBER 1997 1 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) THE S O IT II AFRICAN JOURNAL OF SPORTS MEDICINE VOLUME 4 NUMBER 4 NOVEMBER 1997 Editors P rofT D Noakes D r MP Schivellnits Editorial Board Dr M E Moo I la D r P de Jager Dr J Skowno D r P Schwartz P ro f R Stretch D r C de Ridder P r o fB C Andrew s Dr E W Derman Mr R H Farman Dr M Mars D r C A Noble International A dvisory Board Lyle J Micheli Associate Clinical Professor o f O rthopaedic Surgery Boston, USA Chester R Kyle Research Director, Sports Equipment Research Associates California, USA P r o fH C Wildor Hollmann President d es Deutschen Sportarztebiuides Kohi, West Germany H oward J Green Professor, Department o f Kinesiology Ontario, Canada G eorge A Brooks Professor, Department o f Physical Education California, USA Neil F Gordon Director, Exercise Physiology Texas, USA Edmund R Burke Associate Professor, Biology Department, University o f C olorado Colorado, USA Graham N Smith Physiologist Glasgow, Scotland CONTENTS Guest Editorial - Sport Psychology C Gahwiller Concepts in pain mechanisms and management PA Scott Personality style and mood states associated with a negative addiction to running SJ Anderson, CJ Basson, C Geils, R Farman Psychological preparation for a low intensity marathon squash event CJ Basson 6 12 Psychological, kinematic and psychophysical differences between overground and treadmill running 17 S Olivier The Editor The South African Journal o f Sports Medicine PO Box 115, Newlands 7725 PRODUCTION Andrew Thomas PUBLISHING Glenbarr Publishers ee Private Bag X I 4 Parklands 2196 Tel: (O il) 442-9759 Fax: (O il) 880-7898 ADVERTISING Marika de Waal/Andrew Thomas REPRODUCTION Output Reproduction PRINTING INCE The views expressed in individual articles are the personal views o f the Authors and are not necessarily shared by the Editors, the Advertisers or the Publishers. No articles may be reproduced without the written consent o f the Publishers. 2 SPORTS MEDICINE NOVEMBER 1997 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) Concepts in Pain Mechanisms and Management PA Scott Department of Human Movement Studies, Rhodes University ABSTRACT The challenge to understand the som ewhat nebulous concept o f pain has been around since A ristotle classi­ fied it as the "passion o f the soul", and most a thletes will have experienced pain to a lesser or grea ter degree at som etim e in their career. Although there is a clear physiological com ponent as the body responds to som e noxious stimuli, as soon as the pain sensation is ‘felt' it becom es a personalised experience affected by a multitude o f psychosocial fa c ­ tors. The interaction between physiological and psycho­ logical factors results in pain being a com plex multi dimensional phenomenon which is not easily under­ stood. INTRODUCTION Most p eople will experience pain at som e stage hi (heir lives, but no sp ecific group is more likely to experience pain than athletes. C lich es like “ No pain, 110 gain” and “ pain is temporary, pride is forever” arc part o f a drive to push to, and beyon d, a th letes’ phys­ ical lim its. Pain has been iden tified as an integral part o f the ath letic experien ce and the phenom enon o f pain is therefore o f great relevance to all involved in sporting activities. John Heil o f the United States, p robably one o f the m ost prolific w riters on the psy­ ch ologica l a sp ects o f sport injuries, has identified pain as a “ com p lex m idtidim ensional p hen om en on ” w hich con stitutes an everpresent challenge to the athlete '. The exp erience o f pain a ssociated with phys­ ical perform ance falls into three broad categories: Pain as a residt o f traiunatic injury; Paul as the result o f excessive physical d em ands; Pain suffered during the rehabilitative period as the residt o f one o f the above. As athletes arc pushed to their physical lim its their b o d ie s are under continual stress, the residt being that fatigue, d iscom fort and ‘ real pain’ are c o m ­ mon experien ces. Thus an understanding o f the c o n ­ cep t o f pain and its p ossible effects on perform ance warrants system atic investigation. Insight into the theoretical background o f the com plexity o f the p h e ­ nomenon o f pain sh oid d help in understanding why athletes respond so differently to painful experiences. Concept of Pain It is interesting to note that Aristotle is reported to have cla ssified pain as the ’’passion o f the sou l” ; som e d efin ed pain as an unpleasant subjective exp e­ rience familiar to everyone, w hile Heil* iden tifies pain as “ ... a com p lex biop sych ological p h en om en on ” . The vast array o f d efinition s o f pain, together with varying exp erien ces o f pain, give rise to the question as to w hether pain is a physiological or psychological phenom enon. Address for corespondence Department of Human Movement Studies Rhodes University GRAHAMSTOWN T h e identification o f ‘pain nerves’ in the 19th ce n ­ tury led to the classification o f pain as a sensation and a recognition o f pain as a com p lex p ercept; a c o m ­ bination o f sensation, and affect, not just a sim ple sen sory ex p erien ce m ed ia ted through p eriph eral pathways alone: MerskV’ wrote that, “ Pain is a m ix­ ture o f sensation and em otion ” . In oth er w ords pain is neither a purely physiological nor purely p sych o­ logical phenom enon, but rather a com p lex interaction o f both dim ensions. However, to gain a full understanding o f this ch a l­ lenging concept it is necessary to first d iscu ss the physiological com pon en ts in isolation before a d d re s s ­ ing the psychological factors to be con sid ered , and finally take an overview’ o f the interaction o f the two. P hysiological Factors or the N ociceptive System H eiP d escrib es the nociceptive system as a su b-divi­ sion o f the central nervous system w hich together with chem ica l m ed iators is responsible for the sensa­ tion and transm ission o f pain. Pain is triggered by the activation o f two identifi­ able types o f receptors, viz the inechanorecep tors and the polym odal n ociceptors. The inechanoreceptors respond to m echanical stim uli (ic external physical p ressure) while the polym odal receptors respond to therm al and chem ica l stiim di and generally have a slow er and longer lasting transm ission, in which the receptors continue to fire for a time after the ce s s a ­ tion o f the pain - provoking stim ulus. Having been transduced to neural energy,these im pulses then travel along the afferent pathways o f Lhc peripheral nerves to the dorsal horn o f the spinal cord w hich, accord in g to H c il’, functions as a sensory nerve “ sw itching station” . Here pain, together with other sensory stim uli from the periphery, converge upon com m on sensory neural pathways ascending to the brain. T his is a very sim ilar con cep t to the “ Gate co n ­ tr o l” theory w here M elzack and Wall*, p rop osed that processin g cen tres within the spinal cord may cith er decrease or increase the intensity’ o f pain as a neuio- electrical phenom enon w hich results in the p ercep ­ tion o f relatively lesser or greater pain than that ini­ tially induced. In oth er w ords the pain stim uli are som eh ow m odidated en route to h igher cortical c e n ­ tres. The final p erception o f pain is based upon the sum m ation o f inputs from several areas o f the brain including those responsible for m em ory and em otion. H eiP goes on to report that as a con seq uen ce o f this input from various areas o f pain, even at the stage o f initial aw areness, the p erception o f pain gam s som e o f its m eaning from prior experience together with the present state o f m ind o f the athlete. Once the experience o f pain has registered, the brain sets o f f a cascade o f electroch em ica l events within the n ociceptive system , and this heightened activity within the neurochem ical system s o f the body will have a d ir e ct influence on the su bsequ en t response to any pain-inducing stim uli. SPORTS MEDICINE NOVEMBER 1997 3 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) Figure 1: A schematic representation o f the multidimen­ sional factors which contribute to the personalised experi­ ence o f pain. (Adapted from Scott and Munton, 1995). However, although pain is recognised as being a physio­ logical response, the intensity o f that response is a p h e­ nomenological experience in that it is very specific to the individual and is greatly influenced by psychological fac­ tors. Skulety13 pointed out that while numerous physio­ logical factors give rise to the sensory phenom enon o f pain, psychological components influence the individ­ ual’s reaction to the pain. Norris' supports this argument by stating that th e actual ‘feeling’ or being aware o f pain stimuli is about th e same for everyone, but acknowledges that the overall pain experience varies greatly from indi­ vidual to individual. Psychological Factors Heil3 reports that the instant the individual perceives the pain it becom es predominantly a psychological p h e­ nomenon; stating that perception sets o ff a psychologi­ cally driven chain o f reactive events, all o f which result in a personalised response to the pain. Fine 1 writes o f pain being a ‘private’ experience and it is a well known fact that people respond very idiosyncratically in pa in (til situ­ ations. The m ost basic reasoning behind this is human variability not only in the innate physical and mental make-up o f individuals, but also their past experience, the present situation in w hich the pain is experienced, and finally a cognitive and affective analysis o f how the pain situation could affect future performances. Acknowledging the uniqueness o f individuals, many researches have focused specifically on the personality o f individuals and their response to pain. Petrie10 identified different ‘types’ o f people and their response to the expe­ rience o f pain. There are the “ augmentors” w ho appear to exacerbate the intensity o f pain while “ reducers” are those who do not appear to “ feel” pain as much as oth­ ers. Others have em phasised the cultural influences and argue that the importance o f the social com ponent in the response to pain has led to the characterisation o f pain as a biopsychosocial phenomenon. The main point here is the importance to recognise the network causality: - an athlete’s response to pain will in all probability not depend on any one particular facet, but rather on the interaction o f several personal (both physical and emotional) and social factors. The resulting individual perception o f the pain will clearly affect one’s personal response to the situation. Melzack and Wall5 in their classic paper on the sub­ ject, talk o f pain as being biologically based with psycho­ logical dimensions. Clearly the primary factor must b e physiological in that the noxious stimuli are detected by the sensory pain receptors. However, the affective com ­ ponent is the basis o f the emotional response o f the indi­ vidual, and finally there is the cognitive com ponent which will draw on personal appraisal o f the entire am bi­ ence at the time o f the in jury. Pain Threshold and Pain Tolerance T he actual experience o f pain can b e expressed in t wo distinct forms, viz the threshold o f pain and pain tolerance. The threshold o f pain is the level o f stimulus above which pain will b e experienced11 and is an unlearnt, innate physiological response; ie it involves the stimula­ tion o f th e specialised pain receptors and many are now o f the opinion that this pain threshold is very similar for everyone. On the other hand pain tolerance is the indi­ vidual’s endurance o f the pain and is very m uch more a learnt response affected by various psychological factors, specifically the present situation: therefore the tolerance o f pain varies greatly between people. Clearly a person’s ability to tolerate pain is extremely important in m ost sporting environments, particularly in endurance type activities. Morgan8 reported that the ability to perform successfully in a marathon is governed by the individ­ ual's physiological capacity together with a willingness and capability to endure extreme discom fort and pain. Whitmarsh and Alderm an15 propose that athletes with higher pain tolerance are likely to perform better than those with low pain tolerance; and Heil4 proposes that many o f the greatest sports victories are played out in a field o f pain and injury. Another useful conceptual dichotom y in the phenom ­ enon o f pain is that o f acute and chronic pain. Simply put the former is short term or im m ediate pain, and the lat­ ter is long lasting pain. But is the temporal factor the only difference? Acute pain is generally experienced at the time o f the onset o f trauma. Frymoyer and Waddell2 state that acute pain usually bears a relatively straight forward relation­ ship to the external cause, the intensify o f the stimulus and tissue damage. They report that acute pain is gener­ ally proportional to the physical finding and usually has a specific location and is easily identifiable. While the dominant com ponent o f acute pain is physical, it is gen­ erally accompanied by the affective state o f anxiety as the athlete worries about the severity o f the in jury This is in contrast to chronic pain which appears to b e more diffuse in nature and is far m ore complex. Chronic pain tends to cause continual debilitating discomfort and b ecom e increasingly disassociated from the physical problems; becom ing m ore increasingly associated with emotional distress, depression and a failure to cope. As pain continues, the individual b ecom es preoccupied with the pain and interpersonal functioning is adversely affected. This is com monly m anifested as anger, hostili­ ty and even withdrawal from social contacts. Frymoyer and Waddle3 warn that chronic pain may b ecom e a pro­ gressively self-sustaining condition, eventually becom ing more psychological than physical. This argument is sup­ ported by Heil4, who points out that pain becom es chron­ ic when complaints o f pain b ecom e a way o f expressing distress and when inconsistencies in painful behaviour are noted. Assessm ent o f Pain Due to the com plex and highly subjective nature o f pain it is difficult to gain a tangible, objective assessment o f the amount o f pain being experienced b y the individual. T he sim ple self-report scale o f “ 0 to 10” (see Figure 2) is 4 SPORTS MEDICINE NOVEMBER 1997 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) versatile and allows pain to b e assessed across a variety o f situations, and is accepted as a valid indicator o f pain intensity. With regular use o f the scale one can gain m ea­ sures o f ‘average training pain’ and ‘worst injury pain’, and so identify specific factors or situations that lead to an increase or decrease in pain. According to Heil3 this line o f enquiry and attempt to quantify the level o f pain will in tim e provide insight into the physical challenges the athlete faces in training, com peting and rehabilita­ tion; the training and rehabilitating are o f particular importance as they give one a perspective o f the athlete’s coping mechanism s and how effectively these are work­ ing. PAIN PERCEPTIO N 0. NORMAL 1. 2. UNCOMFORTABLE 3. 4. VERY UNCOMFORTABLE 5. 6. PAiNFUL 7. 8. VERY PAINFUL 9. 10. EXTREMELY PAINFUL Figure 2: The Pain Scale. Pain Management It is not the focus o f this paper to discuss the surgical or pharmacological measures which may b e necessary to repair damaged tissue, but rather the focus is on the per­ sonal management o f pain; in other words the psychologi­ cal coping strategies. Weinstein14 wrote o f pain as having a unique unpleasant affective quality, and in order to cope with this experience an athlete has to mobilise special resources. It should b e noted that the experience o f pain is a warning signal and it is important to realise that ‘pain’ is real to the sufferer; no matter how minor the injury appears to be, i f an athlete complains about pain it is indicative that something is wrong. Millar7 argues that all pain is felt in the mind - even the pain o f a sprained ankle. Psychological problems asso­ ciated with the experiences o f pain will hinder athletic per­ formance and should not b e taken lightly. Athletes com ­ plaining o f pain must be listened to and appropriate steps must b e taken. There is a need to assess both the physical area as will as possible underlying factors. Heil3 proposed a “pain-report attentional matrix” wherein cognisance must b e taken o f the sporting activity, the cause o f the pain, plus the individual. In his conceptual schem e he identifies four broad classes o f pain coping methods. These are defined by whether the athlete “ focuses on” or “ focuses away” from the pain and the activity simultaneously, or, whether there is a separation o f attention between the activity and the pain. The coping strategy selected will depend on the individual, the situation, the activity and the interaction o f these components. A t the tim e o f the occurrence o f an injury the athlete experiences pain, fear and anxiety'. An injured athlete is likely to b ecom e lost in emotional and irrational think­ ing. It is therefore essential to create a sense o f calmness and security by firstly identifying th e site, cause and extent o f injury; then shifting focus from the pain to thoughts o f injury management, rehabilitation and posi­ tive outcome. Explain the mechanism s o f injury and the treatment thereof, establish a rehabilitation programme, identifying pain as a routine aspect o f rehabilitation. In other words acknowledge the problem and direct atten­ tion to the solution, keeping the athlete realistically aware o f the situation and involved in the rehabilitative routine. T he emphasis should b e on a safe and speedy return to action. It is also important to help the athlete differentiate between “ acceptable or routine pain” and “ dangerous pain” . The former must b e seen and accepted as ‘normal’ in the training, competing or rehabilitation programme and something which can and must b e tolerated to a cer­ tain degree, while “ dangerous pain” must b e recognised as such and appropriate steps taken to eliminate the cause. Conclusion Pain is clearly a com plex multi-dimensional phenom e­ non incorporating sensory, affective and cognitive com po­ nents and it is only with a greater understanding o f the general concept plus the individuality o f response to pain that one will b e better able to assist athletes to cope with PAIN. References 1. Fine P G (1 9 9 3 ). The biology o f pain. In J Heih Psychology o f sport injury. Champaign. Human Kinetics Publishers, pp 269-280. 2. Frymoyer J W and Waddle G (1 9 9 1 ). A cu te and chronic pain in M H Pope, GBJ Andersson, J W Frym oyer and D B Chaffin ( e d s ) . Occupational low back pain assessment, treatment and prevention. St.Louis: M oskey Year Book, pp 71-94. ( 3. Heil J (1 9 9 3 ). Psychology o f Sport Injury. Champaign, IL. Human Kinetics Publication. 4. Heil J (1 9 9 5 ). Sports Medicine Injury Checklist. Sports Medicine Update, 10(3):16-19. o. Melzack R and Wall P D (1 9 6 5 ). Pain mechanisms: a new theory. Science, 150:971-979. 6. Mersky H (1 9 7 3 ). The perception and measurement o f pain. Journal o f Psychosom atic Research, 17:251-255. 7. Millar A P (1 9 9 5 ). The psychology o f sport. Sports Medicine update, 10(3):13-15. 8. Morgan W P (1 9 8 1 ). Psycho-physiology o f s e lf awareness during vigor­ ous physical activity. Research Quarterly for Exercise and Sport, 52(3)-.385-427. 9. Norris CM (1 9 9 3 ). Sports injuries: diagnosis and management fo r phys­ iotherapists. London. Butterworth-Heinemann. 10. Petrie A (1 9 6 7 ). Individuality in pain and suffering. Chicago: The University o f Chicago Press. 11. Scott PA and Mackenzie B (1 9 9 2 ). Pain and how workers handle it. Safety Management, 1(3):1 4-16 . 12. Scott PA and Munton L (1 9 9 5 ). A th letes’ perception o f pain. Paper pre­ sented at 6th South African Sports Medicine A ssociation Congress. Durban South Africa. 13. Skulety FM (1 9 8 4 ). The management o f the chronic pain patient: clini­ cal considerations. J oumal o f Orthopaedic and Sports Physical Therapy, 5 ( 6 ) -.305-307. 14. Weinstein J (1 9 9 1 ). Neurophysiology o f pain. In TG Mayer, V Mcx>ney and R J Gatchel (e d s ): Contemporary conservative care fo r painful spinal disorders. Philadelphia: Lea and Febiger, p p 67-73. lo . Whitmarsh B G and Alderman R B (1 9 9 3 ). The role o f psychological skills training in increasing athletic pain tolerance. The Sport Psychologist, 7:388-399. Q SPORTS MEDICINE NOVEMBER 1997 5 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) Personality Style and Mood States Associated with a Negative Addiction to Running SJ Anderson MSc (Clin Psych) (Natal) CJ Basson BEd. (UNISA), UED (Natal), MA (Psych) (Natal). C Geils BA (Hons), and R Farman BPhil MEd (Birmingham) ABSTRACT Personality subtypes and mood states were investigated in addicted and non-addicted runners (n = 4 9 ), and con­ trasted with a control group o f non-exercisers (n = 3 4 ). Runners were assigned to a non-addicted and addicted group on the basis o f Hailey and B ailey’s Negative Addiction Scale, and all study participants were then asked to complete a biographical questionnaire, the Profile o f Mood States (POMS) and Millon Clinical Multiaxial Inventory (M CM I). Results revealed a group difference on the Vigor subscale o f the POMS, while the Total Mood Disturbance score discriminated the non-addicted runners and the non-exercisers. Although a trend towards dis­ turbed mood was evident for the addicted runners, this failed to reach statistical significance. The MCMI revealed a number o f group differences in personality subtypes (Avoidant, Dependent, and Schizotypal), as well as for more transient clinical symptomatology (Anxiety and D ysthym ia). Taken together, the results suggest that high addiction to running is associated with discrete personality and mood profiles that may contribute to diminished health benefits o f running and exercise. The study findings are dis­ cussed with reference to personality theory, and implica­ tions for further research in this area. INTRODUCTION It is well known that som e athletes engage in intermit­ tent and moderate running, whereas others run with per­ sistent and stubborn regularity. For the latter, running becom es an activity o f central importance in their lives to the exclusion o f all other recreational activity, suggestive o f the properties o f addiction. Sachs and Pargman,'1 define exercise addiction as: Psychological a nd/or physiological dependence upon a regular regimen o f physical activity. Additionally exer- cise addiction is characterized by recognizable with­ drawal symptoms when the need to exercise remains unfulfilled after 24 - 36 hours. These withdrawal symp­ toms may encompass both psychological and physio­ logical factors, including feelings o f irritability, tension, guilt, uneasiness, bloatedness, muscle twitching and discomfort (p. 143). Various terms are used in the literature to describe this phenomenon, including running addiction, obligatory run­ ning, running dependence, and compulsive running. The origins o f the term running/exercise addiction date back to the late 1960s,2 and have resulted in a distinction being A dd ress for C orrespondence: SJ Anderson, Departm ent o f Psychology, University o f Natal (PMB), P Bag X 01 , Scottsville, 3209. Tel: (03 31 ) 260-5372 email: andersons@psy.unp.ac.za m ade between positive and negative addiction. This development resulted from a need to emphasise the detri­ mental physical and psychological effects resulting from a dependence on running. The transition from positive to negative addiction has been linked to a time dimension,319 and is described by Sachs2 as resulting from running shift­ ing from being an important but well considered aspect o f the runner’s existence, to one that controls the runner's life. The major result is that choices in other important areas o f the individual’s life are eliminated and /o r reduced.' Such runners were found to adhere to a regular pattern o f running behaviour and maintain it despite debilitating pain, injuries and medical advice, and in pref­ erence to social, familial or professional responsibilities.2 Noakes and colleagues, and d e Coverley Veale cite cases o f runners with diagnosed myocardial infarction, who ignored severe symptoms while running, and as a result died from myocardial infarcts during marathon events.”0 A further refinement following on the observed relation­ ship between exercise addiction and anorexia nervosa, has been introduced by Veale' who distinguishes between primary and secondary exercise addiction. The latter he sees as suggestive o f d ie presence o f an eating disorder, whereas the former he suggests not being related to any discrete mental disorder. Etiological explanations have included both physiolog­ ical and psychological perspectives. The former proposes that increases in levels o f plasma B-endorphin released into the central nervous system in response to the stress o f exercise, with subsequent m ood enhancing and anal­ gesic effects, are responsible for the development o f addiction;” 10 however, the findings are equivocal.0 What has not been established, is why only some individuals becom e addicted to running. Psychological investigations into the personality characteristics associated with run­ ning and running addiction in general have been conduct- ed.2~ 2sai Recent overviews o f the relationship between sport and personality have sketched the long history o f dispositional approaches to the study o f this relation- sliip.30,3132 This trend has been questioned and the use o f more interactional paradigms has been strongly advocat­ ed.*131 Cox is more optimistic especially if use is m ade o f relevant theory as well as multivariate statistical m ethod­ ology such as regression analyses."' Most studies have made use o f inventories such as the 16PF, the Minnesota Multiphasic Personality Inventory7 (MMPI), the Eysenck Personality Questionnaire (EPI) and the Profile o f Mood States (POMS) to measure personality7 dimensions, but few' have used the theoretical underpinnings o f the ques­ tionnaire to inform their research questions.30 31 H Vealey, in her review7, concludes that regardless o f the fact that there is evidence to suggest that runners’ personalities are characterised by introversion, stability, low7 anxiety, self-sufficiency, high self-esteem and imaginativeness, no prototypical athletic personality7 type has yet b een defined.31 6 SPORTS MEDICINE NOVEMBER 1997 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) mailto:andersons@psy.unp.ac.za With regard to personality and addiction, Steinberg and Sykes" cite Pargman and Balter’s review o f the liter­ ature, suggesting that there may b e underlying similari­ ties in the psychological profile o f addicts generally. The idea o f an addictive personality' type has, however, not received much support from current research. Nonetheless, researchers continue to investigate the per­ sonality profile o f the addicted exerciser, and more par­ ticularly the runner. - 11 - 27 28 A specific personality' attribute aasociated with addic­ tion, and suggested to b e com mon to addicted exercisers and anorexics, is the obsessive-compulsive and anxious constellation.12 3 Compulsive individuals may b e attracted to running as a coping st rategy for stress because o f the repetitive routinized nature o f the activity and its capacity' to facilitate control over mood states as well as physical appearance and functioning.7 Although some research supports a higher incidence o f obsessive-compulsive fea­ tures and anxiety' in addicted runners, other evidence sug­ gests that runners present with diverse personality traits without significant pathology.301331 Blumenthal, O Toole and Chang found that addicted runners scored within d ie normal range on the MMPI, and in addition found that even the m ost severely addicted runners did not exhibit significant psychopathology as measured by the MMPI.13 They argued that a disease model o f the addicted runner w'as misleading and unnecessarily pejorative. Yates, Shisslak, Allander and Crago found the obligatory and non-obligatory runners in their study to have more simi­ larities than differences, although obligatory runners tended to prefer social isolation.29 Others have found male obligator}7 runners to b e more motivated by achievement and a need for recognition.2” Thus som e trends have emerged in recent research although no theoretical con­ ceptualisation o f personality or running addiction has informed these studies. A further area o f interest to the present study is the rel­ atively well established finding that exercise results in an increase in positive m ood states and a decrease in nega­ tive m ood states (e.g. anxiety and depression), and that physical exercise is associated with positive nioocl.H lsln:11 Chan and Grossman17 cite Morgan and Pollock as noting that, compared to non-exercisers, runners w'ere found to be lower in depression, fatigue, confusion, and tension as measured by' the POMS, and hypothesized that these mood differences w'ere a positive consequence o f running. Millon2'’ cites Morgan as finding similar evidence, and naming this constellation o f scores the ‘Iceberg profile’ by virtue o f it’s visual graphic configuration. This finding may also indicate that individuals with more positive mood characteristics choose to exercise, and that addicted run­ ners d o not derive the positive mood benefits from run­ ning. Purpose of the study In view' o f the uncertain findings in the literature 011 the relationship between personality and mood states and running addiction, this study aimed to examine whether negatively addicted runners are characterized by a com ­ pulsive personality' style (as measured on the Compulsive scale o f the MCMI). Furthermore it was proposed to examine w'hether there are any other personality differ­ ences that distinguish negatively addicted runners from non-addicted runners, and whether addicted runners tended to demonstrate greater mood disturbance 011 the POMS. Further investigation into these variables could shed light 011 management and intervention programmes lor use by clinicians, coaches, and athletes. Theodore MUlon’s Biosocial Learning Theory' o f person­ ality W'as used to conceptualise personality functioning.2420 Personality is defined by Millon as: A d istinctive configuration o f interlocking perceptions, feelings, thoughts and beha\iors that provide a tem ­ plate and disposition for maintaining psychic viability and stability (p. 643).2,1 Pathology is viewed as occupying the extreme end o f a continuum o f normal behaviour, and differentiated by the following three features: functional inflexibility and rigid­ ity; a tendency' to foster vicious or self defeating circles o f behaviour; and tenuous structural stability under condi­ tions of stress.20 Each personality scale is thus construct­ ed on the basis o f a personality prototype rather than a syndrome or type against which individuals are compared for their closeness o f fit.2' Each prototype is m ade up o f a small and distinctive group o f primary' attributes that persist over time” (p. 673), that are utilised by the indi­ vidual to ensure predictability, viability, and control in adaptive contexts. These prototypes have been opera­ tionalized in the Millon Clinical Multiaxial Inventory (MCMI) and its revisions (MCMT-II and MCMI-III). METHODOLOGY Subjects An opportunity sample was secured from a list o f m em ­ bers o f a university athletics club and a list o f those run­ ners from tlie club who had com pleted the Comrades Marathon in that year. The remaining runners were drawn from two running clubs in KwaZulu-Natal on the basis of their commitment to a regular running pro­ gramme. The final sample consisted o f 49 runners (33 males and 16 females), ranging in age from 19 to 63 ( = 25.84 years). A comparison group o f 34 non-exercisers (14 males and 20 lemales) was selected by opportunity sampling, from a university staff/student population. They ranged in age from 19 to 50 ( = 2 3 .2 years). They were selected on the basis that they d id not participate in regular exercise (i.e. two or more times a week). Exercise was either sporadic and/or primarily o f a low intensity, e.g. casual walking or yoga- x The runners were assigned to one o f three groups on the basis o f their scores on the Hailey and Bailey Negative Addiction Scale (see results section).13 X INSTRUMENTS Information Questionnaire This instrument included biographical information and questions relating to the different types o f exercise the subjects typically engaged in, the intensity' frequency, duration o f such exercise, and motives for exercising. The Negative Addiction Scale The Negative Addiction Scale (NAS) was designed by Hailey and Bailey111 to objectively quantify' the psychologi­ cal aspects o f Morgan’s concept o f ‘negative addiction’. It is a 14-item questionnaire that yields a single addiction score ranging from 0 (low) to 14 (high). Furst and G em ion e1 found that although the maximum score on this scale is 14, the m ost highly addicted runners in their study had a mean addiction score o f 6.38. No research has SPORTS MEDICINE NOVEMBER 1997 7 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) been done to identify the point on this scale at which the onset o f addiction to running occurs. ’ T he psychometric properties o f this scale (ie. validity and reliability) are not mentioned, but it was, however, the m ost available scale at the time o f the research. Thornton and Scottr- have sub­ sequently used this scale with some success. Profile o f Mood States (POM S) T he POMS-’ was designed to measure transient and fluc­ tuating m ood states,10 although it is regarded by some researchers,3132 as a dispositional measure rather than a state measure o f personality. The scale is noted by Morris32 as having been extensively used in at least 12 studies suc­ cessfully distinguishing levels and types o f athletic per­ formance. It has 65 items that consist o f adjectives describing various feelings and moods, each o f which is rated on a 5-point intensity scale. The results yield six dimensions o f mood: tension/anxiety; depression/dejec­ tion; anger/hostility; vigor/activity; fatigue/inertia; confu­ sion/bewilderment. Vigor is subtracted from the total o f the other five m ood states to provide a Total Mood Disturbance score, with a high score indicating poor adjustment. This Total M ood Disturbance score is highly reliable because o f the inter-correlations among the six primary POMS factors, and reliability o f the scales ranges from 0.65 to 0.90.23 MiUon Clinical Multiajdal Inventory (MCMI) T he MCMF4 consists o f 175 items, from w h i c h 20 scales are derived. Scales 1 - 8 reflect the persistent, ingrained, enduring and pervasive personality characteristics or ‘per­ sonality style’ o f an individual. Scales S, C, and P reflect a greater level o f pathology in the p e r s o n a l i t y pattern, while scales A through PP are clinical syndrome dimensions that reflect current symptom states o f a transient nature (i.e. symptoms that fluctuate over time in response to the impact o f stressful situations). These latter symptoms typ­ ically exacerbate the basic personality style. Test-retest reliability indicates that the overall stabili­ ty o f the MCMI scales is reasonable, with an average around 0.80 for both the personality patterns and the pathological personality scales.24 Stability for the clinical syndromes, which are more transient and will inevitably fluctuate, is lower at 0.65.24 Concurrent validity was sup­ ported by correlational data between the MCMI scales and three comparable diagnostic inventories, including theM M PI. B a s e r a t e scores provide uniform cutting lines: the BR75 cutting line is drawn to identify the presence o f a trait or disorder. The BR85 cutting line represents the most prominent s\ndronie(s). RESULTS (a ) Negative Addiction Administration o f the NAS to the runners revealed a wide spread o f scores (range 1-13; = 3,82, SD = 2.45). Following the findings o f Hailey and Bailey, the runners Were assigned to one o f three groups on the basis o f their level o f negative addiction on the NAS (see Table 1). The high addiction subgroup comprised 11 runners (6 male, 5 female) scoring 6 or more on the NAS (Hailey and Bailey note that the mean addiction score for the m ost highly addicted runners in their sample was 6.38; SD = 2.61). T he low7 addiction subgroup consisted o f a group o f seven male and three female runners, with negative addiction scores o f 4 and 5. T he non-addicted subgroup (21 male, 6 female) w ere identified on the basis o f their NAS scores between 1 and 3. A median split technique o f the Negative Addiction Scale scores (median = 3) supported this group­ ing. Table I. Sub-Groupings o f Runners on the Negative A ddiction Scale (n 4 9) Subgroup n mean SD Range H igh a ddiction 11 6.36 1.27 6-13 Low7 addiction 10 4.10 0.32 4-5 N on-addicted 28 2.21 0.75 1-3 A series o f one-way ANOVAs failed to reveal any group dif­ ferences between the high and moderate addicted run­ ners on any o f the variables, and so these two groups were collapsed into a single group (addicted runners; = 3.80; SD = 2.46) for the remainder o f d ie statistical analysis. ( b ) Mood States T h e results for the three groups on the Profile o f Mood States (POMS) are summarized in Table II (groups col­ lapsed for sex) and Figure 1. Table II. POMS Raw7 Scores (and SDs) for the A dd icted and N on-A ddicted Runners, and the Non-Exercisers. Scale addicted non-addicted non-exercisers runners runners Tension 9.90 (6.08) 7.78 (4.17) 10.12 (5.48) Depression 8.90 (7.61) 5.15 (4.26) 10.57 (9.68)* Anger 8.65 (7.47) 5.07 (3.80) 10.57 (9.68)* Vigor 21.00 (4.88) 30.96 (3.03) 14.24 (5.90)** Fatigue 7.20 (5.84) 4.81 (4.10) 8.91 (5.43)* Contusion 7.80 (6.01) 6.07 (3.35) 8.36 (5.17) TMD 21.15 (25.01) 7.93 (14.56) 33.59 (28.36)* significant group difference *p <.05; **p <.001 TMD = Total Mood Disturbance A significant difference between the non-exercisers and non-addicted runners, emerged on the Anger (F = 3.53, p < 0.05), Depression (F = 3.76, p < 0.05), Fatigue (F = 4.73, p < 0.05) and Total M ood Disturbance (F = 8.8 1, p < .00 1) scores o f the POMS (Scheffe’ post-hoc, p < .05). The Vigor subscale also revealed a significant difference (F = 18.76, p < 0.001), between the runners as a w7hole, and the non-exercisers (Scheffe post-hoc). In order to assess the clinical significance of the POMS scores, the groups were compared for the presence o f scores (excluding Vigor), above the clinical cut-off point o f T > 60. Results revealed a significant group differences using Chi-squared (F = 9.16, p < 0.02), with the non-addicted runners showing less overall emotional disturbance than the non-exercis­ ers. T he POMS Vigor scale w7as examined separately (it is the only positive mood state), and show7ed that the non­ exercisers reported less vigour than the addicted and non-addicted runners (F = 13)21, p < 0.001). Since the scoring o f the POMS is not adjusted for sex, investigation o f potential gender differences was m ade using a two-way ANOVA (group by sex); however, this failed to reveal any 8 SPORTS MEDICINE NOVEMBER 1997 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) Table III. MCMI Base Rate (BR) Means and SDs for the Addicted and Noil-Addicted Rnniiers, and tlie Non-Exercisers. Personality patterns addicted lion-addicted Scale runners runners non-exercisers Schizoid (1) Avoidant (2) Dependent (3) Histrionic (4) Narcissistic (5) Antisocial (6) Compulsive (7) Passive- 39.29 (25.53) 40.80 (21.31) 68.52(23.41) 66.71 (20.46) 63.52 (13.71) 56.76 (20.67) 65.71 (17.34) 26.25 (17.05) 23.54 (19.12) 50.14 (28.24) 73.04 (15.69) 70.79 (15.37) 62.00 (20.66) 64.50 ( 9.30) 33.85 (23.28) 28.71 (21.60)* 48.97 (26.56)* 74.26 (24.86) 69.56 (20.83) 58.24 (20.01) 64.06 (11.38) aggressive (8) Schizotypal (S) Borderline (C) Paranoid (P) 36.42 (24:42) 56.48 (10.03) 57.52 (16.17) 60.38 (17.85) 30.75 (14.20) 40.39 (20.30) 46.75 (18.54) 63.43 (13.15) 39.53 (21.21) 44.71 (17.56)** 54.91 (19.22) 56.85 (19.46) Clinical syndromes Anxiety (A) Somatoform (II) Hypomanic (N) Dysthymic (D) Alcohol Abuse (B) Drug Abuse (T) Psychotic Thinking (SS) Psychotic Depression (CC) Psychotic Delusions (PP) 74,67 (15.28) 73.33 (14.94) 47.95 (30.60) 64.52 (15:33) 46.57 (19.64) 56.00 (18.69) 51.38 (9.98) 41.33 (17.27) 52.76 (18.48) 59.93 (21.67) 68.44 (21.82)* 65.32 (17.57) 68.12 (14.62) 50.79 (21.38) 46.85 (28.50) 49.39 (18.72) 62.15 (21.02)** 47.29 (13.03) 47.03 (21.07) 61.79 (14.31) 56.06 (24.31) 41.46 (21.73) 44.24 (17.86) 34.82 (15.89) 40.47 (18.85) 49.89 (20.21) 47.29 (22.24) profile differences between males and females. ( c ) Personality and psychopathology Group data for the MCMI basic personality profiles and clinical symptom scales are shown in Table III and Figure 2. A series o f one-way ANOVAs revealed significant group differences for two o f the personality scales (Avoidant, Dependent, and Schizotypal), as well as for two o f the clin­ ical syndrome scales (Anxiety and Dysthymia). Post-hoc group comparisons (Scheffe, p < .05) showed that the addicted runners differed significantly from the non-addicted runners on the Avoidant (F = 4.29, p < .05), Schizotypal (F = 5.56, p < 0.05), and Anxiety (F = 3.27, p < 0.05) scales, whereas the scores o f the Dysthymic scale showed that the non-addicted runners were less depressed than both the addicted runners mid non-exer­ cisers (F = 4.92, p < 0.01). In order to ch eck for the pres­ ence o f clinical pathology on the MCMI, subject profiles addicted ----------- non-addicted non-exercisers Figure 1. POMS scores for addicted and non-addicted runners, and non-exercisers. were examined for the presence o f personality and symp­ tom dimension scores at or above the clinical cut-off base rate o f BR75. Although there were a number o f individ­ ual cases in which d ie cut-offs were exceeded, diere were no overall group differences suggestive o f pathology for the clinical syndromes (F = 3.75, p < 0.20); pathological personality scales o f Schizotypal, Borderline and Paranoid (F = 1.09, p < 0.20); or the basic personality scales (/■’ = 3.86, p < 0. 1 0) using Chi-squared. To further investigate personality differences between groups, one-way (group) and two-way (group by sex) AiYOVA’s were run using the MCMI scales that were sug­ gested in a four fold factor matrix by Millon.24 Significant group (F = 4.12, p < 0.05) differences emerged for Factor 1 with a post hoc Schelfe (p < 0.05) analysis indicating that the addicted runners were more emotionally labile and depressed than the non-addicted runners. Factor 1 is described by Millon24 as “ depressed and labile emotional­ ity expressed in affective moodiness and neurotic com ­ plaints” (p. 49). The scales with maximal positive load­ ings (> 0.80), on diis factor were Borderline, Dysthymia, Anxiety, Psychotic Depression, Passive Aggressive and Histrionic. The Avoidant, Schizotypal, Hypomanic and Psychotic Thinking scales loaded between 0.40 and 0.80. The Compulsive scale loaded significandy negatively on Factor 1 (-0.747). Group by sex (F = 2.644, p = 0.077), and group (F= 2.963, p = 0.057), differences approached sig- significant group difference *(p<.05) **(p <.01) 80 70. 8 60. 8 50- 10 MCMI scales ----------- addicted -----------non-addicted ............ non-exercisers Figure 2. MCMI Scores fo r the A d dicted and Non- A ddicted runners, and Non-Exercisers, a nificance on Factor 3. The addicted male runners tended to be more asocial and avoidant, and low on sociability and self confidence. Factor 3 is described by Millon-’ as a “ core pattern o f schizoid behavioural detachment and thought SPORTS MEDICINE NOVEMBER 1997 9 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) with positive loadings on scliizoid, avoidant and psychot­ ic thinking” (p. 50). Sex differences approached signifi­ cance (F = 3.681, p = 0.059) on Factor 4, with females showing a trend o f greater social restraint than d ie male runners. There were no significant group or group by sex differences on Factor 2, (this factor is related to paranoid diinking and behaviour with associated grandiosity and suspiciousness). The addicted runners thus tend to be emotionally labile, and this is associated, particularly in the male runners, with low sociability and low s e lf confi­ dence. In order to identify which MCMI personality scales b est predicted group membership for addicted and non- addicted runners, a Discriminant Function Analysis (DFA) was performed. This was preceded by a factor analysis (n = 49) in which four factors were extracted (varimax rota­ tion); these were: Factor 1 - Somatization (0.92), Dependent (0.84), Borderline (0.76), Anxiety (0.75), and Antisocial (-0.79); Factor 2 - Drug Abuse (0.87), Alcohol Abuse (0.82), Hypomania (0.73), Passive-Aggressive (0.70), Histrionic, and Compulsive (-0.71); Factor 3 - Schizoid (0.88), Avoidant (0.79), Schizotypal (0.72), Psychotic Thinking (0.66), and Histrionic (-0.62); while Factor 4 was comprised o f Psychotic Delusions (0.97) and Paranoid (0.68). An insufficient number o f variables left in the correlation matrix excluded d ie identification o f any further factors. Three o f these factors (namely Factors 1, 3 and 4), suc­ cessfully discriminated the addicted and non-addicted runners (respective Wilks’ lambda were: .84 for Factor 3 (p < 0.005); 0.80 for Factor 1 (p < 0.006); and 0.78 for Factor 4 (p < 0.009). Canonical discriminant function c o ­ efficients were 0.417, 0.901 and -0.44 for d ie respective factors F 1, F3 and F4. T he results o f the DFA found that these factors were successful in classifying 76.2% o f die addicted runners and 71.4% o f the non-addicted runnel's. T he overall correct classification was 73.47%. A stepwise multiple regression analysis was also per­ formed wil.li scores from d ie NAS as d ie dependent vari­ able,. The first predicted variable was duration o f exercise, which accounted for 21% o f die NAS variance (F = 11.4, p < 0.01). Witii the addition o f d ie Schizotypal Personality subscale from die MCMI, diis percentage increased to 31% (F = 9.3, p < 0.001). Finally, d ie addition o f a tiiird predictor, namely sex, accounted for 40% o f the overall variance o f d ie NAS scores (F = 8.8, p < 0.001). Finally, in order to identify which variables b est pre­ dicted Schizotypal Personality, an additional stepwise multiple regression analysis was performed on d ie data. On the first step, the Total Mood Disturbance score (POMS) emerged as d ie significant variable, accounting for 16% o f the variance (F = 7.9, p < 0.01). The second variable (duration o f exercise) accounted for a further 9% (total variance 24%; F = 6.5, p < 0.0 1). Finally, intensity o f exercise (questionairre data) contributed a further 11 % (total 3 5%; F = 7.2, p < 0.00 1). The 0.05 limits were reached following tiiis last extracted variable. DISCUSSION The identification o f addicted versus non-addicted run­ ners using an arbitrarily-detennined cutoff point on the NAS wras d ie source o f some concern to d ie researchers. However, a median split o f 3 in the current study suggest­ ed that the cut-off point used by Hailey and Bailey10 was compatible widi our own findings, and represents a useful clinical guideline. We would support a view' tiiat further psychometric research is desirable to fully explore d ie validity o f NAS cut-off points. T he m ost prominent feature o f the POMS was d ie ‘ice­ berg profile’ that distinguished the two running groups (high and low addiction) from the non-exercisers (see Fig. 1). This finding is consistent with previous research diat has found that exercising individuals are characterised by higher levels o f energy (Vigor) and lower levels o f em o­ tionality dian non-exercisers.1823 The other POMS findings (i.e. Total Mood Disturbance score, Anger, and Fatigue) also appear to suggest tiiat running is associated with emotional well-being; however, analysis o f non-significant trends in the data suggest that this benefit is maximal for d ie non-addicted runners. It should b e pointed out that considerable overlap exists in the Total M ood Disturbance scores for the addicted and non-addicted runners, as well as for the addicted runners and non-exercisers (see Table ID. The MCMI results lend themselves to useful clinical interpretation. While the results revealed few7 significant overall group differences, the finding tiiat d ie addicted runners w'ere significantly different from both the non- addicted runners and non-exercisers on the Dysthymic scale provides further support for the idea that non- addicted running either promotes positive mood states, or reduces depressed mood. In terms o f personality profiles, our results suggest that running addiction is correlated witii specific personality dispositions, in particular, die Schizotypal personality. This refers to a constellation o f personality dispositions characterized primarily by social detaclunent, a preference for privacy and isolation, and d ie display o f unobtrusive aloofness associated with a ten­ dency towards behavioural eccentricities and low7 self esteem. In addition, there was evidence o f a wrealter correlation between running addiction and Avoidant and Dependent personality patterns. The Avoidant pattern has similari­ ties with d ie Schizotypal disposition in that it is also char­ acterised by social detachment, but witii greater emotion­ al lability, tension, and anger. O f interest to d ie overall aim o f diis study wras the finding tiiat ruiming addiction is not represented by a personality profile that features prominent compulsive dispositions. This dissociation between what on the surface might appear to be com pul­ sive behaviour (i.e. running addiction), and compulsive personality traits, can be interpreted witii reference to Millon’s theory' o f personality as operationalized in the MCMI. To this extent, addicted running appears to be more closely associated with general, rigid, inflexible per­ sonality patterns o f a schizotypal and avoidant nature, in which emotionality has higher prominence than d ie com ­ pulsive personality disposition which Millon describes as being characterised by restrained emotionality. T he sig­ nificant negative correlations (from -.17 to -.38), between the Compulsive sub-scale on the MCMI and all o f the POMS subscales, supports this observation. A further point worth noting concerns d ie largely nega­ tive correlations (from a non-significant -0.03 to a signifi­ cant -0.27), between m ost o f the MCMI clinical scales and the Vigor subscale o f the POMS. This relationship con­ firms the com m on sense expectation that increased feel­ ings o f energy and w'ell-being would be incompatible widi psychological symptomatology in runners. This is partic­ ularly so in relation to the Dysthymia sub-scale, w'here the relationship is significantly negative (r = -0.27, p < 0.01). Other comparisons between the two clinical instruments used in this study are suggestive o f an association between 10 SPORTS MEDICINE NOVEMBER 1997 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) negative m ood state (as measured by the POMS), and a tendency towards specific types o f personality disposi­ tions. For example, our results suggest negative correla­ tions between the Total M ood Disturbance score and the Compulsive personality subtype (-0.31), and high positive correlations (p < 0.001) with Avoidant (0.41), Passive Aggressive (0.55), Schizotypal (0.34), and Borderline (0.49). Somewhat weaker correlations characterized the relationship betw een Total M ood Disturbance and Schizoid (0.28) and Narcissistic (-0.28) personality sub- types. In relation to prerious research documenting the rela- lionsliip between running addiction and personality,1328,29 there is some concordance with the Yates et al. study that found obligatory runners to prefer social isolation.29 However, Blumentlial et al. found no significant differ­ ences between addicted and non-addicted runners using d ie MMPI, and they caution against pathologizing running addiction.™ The findings o f the present study cannot b e directly compared to die Blmnenthal study as the theo­ retical lmderpinnings o f the MMPI and d ie MCMI are somewhat different. The rational for the MCMI is a theo­ ry o f personality, whereas the MMPI is based on a theory o f psychopathology. Furthermore, Millon describes per­ sonality traits as dispositions, and views his sub-types as prototypes against which persons are compared for their closeness o f fit; this has important implicat ions for inter­ pretation o f findings such as ours.20 27 Millon stresses t hat the personality dispositions o f the person approximate the prototype to a greater or lesser degree, rather than being classified into discrete, pathological categories. To this extent, we would concur with Blumentlial and colleagues, that a rigid categorisation o f atliletes into personality and psycliopadiological subtypes is misleading and undeserv­ ing. Specifically, researchers need to contextualize their results and interpretations with reference to the theoreti­ cal underpinnings o f the instruments used. Such an approach, would hopefully lead to increased understand­ ing o f the dynamic psychological processes underlying run­ ning addiction. In terms o f our suggestions for further research, d ie extent to wliich our findings are generalizable to other sports w ould be an important extension. Further studies on d ie utility and reliability o f the NAS also seem wrar- ranted, given that parameters o f running addiction (e.g. stage-like process) have not yet been sufficiently opera­ tionalised, despite pioneering attempts in this regard.2 9,15'-22 Li view* o f die criticisms directed at personality trait r e s e a r c h , d i e use o f more interactional and idio- grapliic research designs, in an attempt to examine the relationsliip between personality and running more specif­ ically, needs to b e implemented. Furthermore multi-dis­ ciplinary and multivariate statistical procedures that include biophysical, psychological and sport skill vari­ ables;^ may b e useful in furthering the knowledge base in this important area o f sport psychology. References /. Coen SP &. Ogles BM (1 9 9 3 ). Psychological characteristics o f the oblig­ atory runner: A critical examination o f the A norexia Analogue Hypothesis. Journal o f Sport and Exercise Psychology, 15,338-854. 2. Sachs ML (1 9 8 1 ). Running Addiction. In M Sacks & M Sachs (E d s.), Psychology o f Running, pp. 116-125. Champaign HI: Human Kinetics. 3. Veals D (1 9 9 5 ). Does primary exercise dependence really exist? In J Annett, B Cripps, and H Steinberg (E d s.), Exercise Addiction: Motivation for Participation in Sport and Exercise. London: BPS. 4. Sachs M L & P a rg m a n D (1 9 7 9 ). Running addiction: A depth interview examination. Journal o f Sport Behaviour, 2, 143-155. 5. Colt EWD, Dunner DL.,.IlaU K & Fieve R R (1 9 8 1 ). In M Sacks & M Sachs (EcLs.), Psychology o f running, pp. 234-245 Champaign IU: Human Kinetics. 6. De Coverley Veale D M W (1 9 8 7 ). Exercise Dependence. British Journal o f Addiction. 82. 735-740. 7. Sacks M H (1 9 8 1 ). Running addiction. A clinical report. In M Sacks & M Sachs (E d s.), Psychology o f Running pp. 127-129.Champaign HI: Human Kinetics. 8. Biddle S i l l &. Mutrie N. (1 9 9 1 ). Psycholoits. International Journal o f Sport Psychology. 26, 233-248. 29. Hites A . Shisslack CM, A llenderJ & Crago M (1 9 9 2 ). Comparing oblig­ atory to nonobligatory runners. Psychosomatics. 33, 180-189. 30. Fields KB, Delaney M &. Hinkle JS. (1 9 9 0 ). A prospective study o f Type A Ijehiavior and running injuries. Journal o f Family Practice, 30, 425-429. 31. van der Auweele Y, de Cuyfjer B, van Mele V & Rzewricke R (1 9 9 3 ). Elite performance and f?ersonality: From description and prediction to diagnosis and intervention. In RN Singer, M Murphy & L K Tennant, (E d s.). Handbook o f Research on Sjx>rt Psycholort Psychology. Champaigns. HI: Human Kinetics. 33. Morris T (1 9 9 5 ). P sychohgiad characteristics and sports behaviour. In T Morris & J Summers, Sports Psychology. Theory, Application and Issues. Brisbane: John Wiley & Sons. 34. Regnier G. S