Aim: Do patients think doctor's dress matters? What is their preference? Does it affect belief in the doctor's effectiveness? Any demographic differences?

• Sample: 475 patients, 5 GP surgeries, Scotland – interviewed after visiting the doctor.
• Procedure: shown 8 photos – 5 of one male doctor, 3 of one female doctor, dressed differently, same pose. Male: white coat over suit · formal suit · tweed jacket + tie · cardigan/sports shirt/casual trousers · jeans + t-shirt. Female: white coat over skirt & jumper · skirt, blouse, woollen jumper · pink trousers, jumper, gold earrings (fewer variations – fewer distinct styles for women GPs).
• Asked "which doctor would you be happiest to see first time?" → score 0–5 per photo; also confidence in ability, would they be unhappy consulting, which looked most like own doctor; closed Qs on dress in general.

• Male: smart suit most popular → tweed jacket + tie ≈ white coat over suit → jeans lowest by far. Female: traditional (jumper + skirt) most popular → white coat. Scores for the woman > scores for the man.
• Age: older patients gave higher scores to white coat (both doctors). Class: social class 1 & 2 (professional/managerial) rated traditional dress higher; male in white coat top for class 1.
• Strongest association = which surgery the patient attended (preferences varied across almost all categories).
• 41% felt more confident in the ability of any one doctor – most for male in suit → white coat; female in white coat → skirt & jumper. 28% would be unhappy consulting with any (male in jeans most chosen). 64% said dress is very/quite important.

Aim: perceived vs actual understanding of medical terms by lower working-class families in Scotland.

• Procedure: researcher joined doctors on rounds for 1 month in obstetrics/gynaecology wards → list of 57 words used by staff/clinic leaflets → doctors crossed out words they wouldn't normally use → 13 words.
• Mothers interviewed: word sounded out, then used in a sentence, mother explained meaning. Responses recorded; 2 doctors scored each independently for accuracy.
• 18 other doctors rated each word: would the average lower working-class mother (a) not understand & say so (b) get it wrong (c) vague/incomplete (d) understand well?
• Compared under-utilisers (didn't attend antenatal care regularly) vs utilisers (did).

• The two scorers' marks closely matched. Utilisers understood better than under-utilisers on 11 of 13 words; utilisers who weren't first-time mums also did better than first-time mums ⇒ experience raises understanding.
• Doctors greatly underestimated understanding. e.g. "breech": adequately understood by 84% under-utilisers & 100% utilisers – doctors predicted adequate in only 22%.
• 3 (technical) terms doctors overestimated. Senior doctors seemed to expect more understanding (too little data to analyse).
• Negative correlation between no. of words a doctor says they use regularly and their expected patient understanding ⇒ doctors use words without expecting patients to know them.

69 patients at a clinic: computerised interview vs paper questionnaire passed to doctor/nurse. Most info given in the computer interview > questionnaire > face-to-face with doctor. ⇒ Patients may not give full, accurate info, esp. if embarrassing/sensitive. + real clinic = high ecological validity; – one clinic, small sample.

• 77 patients with suspected skin cancer chose how to hear their biopsy result: face to face / telephone / letter. Given an info leaflet on pros & cons.
• 1 month later: closed-question questionnaire (how received, leaflet given?, satisfied then & in hindsight?).
• Choices: letter 48% · phone 37% · face to face 11% · combination of two 5%. 89% got it in chosen way (7 couldn't be reached by phone → letter).
• 94% happy with how they got it · 80% would keep same choice in hindsight · only 11% wished for a different one · only 5% would retrospectively choose face to face. No age difference.

131 skin-cancer patients; 32% told by phone, rest face to face. No sig. difference in satisfaction, anxiety or depression between phone and face to face.

• Satisfaction ↑ if: felt prepared · the word "cancer" used · felt "told everything" (severity, life expectancy) · info presented clearly.
• Anxiety ↓ if: prepared · could discuss Qs same day · info level matched needs · emotional needs met (reassurance, willingness to discuss feelings).
• Depression ↓: main factor = info matched to the individual's needs.

• Doctor-centred (directing) style: doctor leads, patient passive; focuses on first symptoms, closed questions (short replies), no discussion, impersonal, aim = find a medical explanation, tells patient what's wrong & what to do.
• Patient-centred (sharing) style: patient active; doctor asks open questions, encourages thoughts & feelings, personal atmosphere, jargon avoided, patient involved in decisions ⇒ joint decision.
• Katon & Kleinman (1980): sharing is best – patient part of decision ⇒ doctor understands problem better, patient more likely to follow advice. Ingelfinger (1980): directing/authoritarian is best for effective medicine – reassures patient.

Aim: compare effectiveness of directing vs sharing consultation styles of a GP on patients' satisfaction.

• One GP, own surgery, 4 months. Random-number generator picked 4 patients per surgery; aged 16–75, any symptom; excluded life-threatening conditions or admin/preventative visits. 200 completed all parts.
• Written consent for audio recording of consultation.
• Cards randomly allocated directing or sharing; prompts for the 5 parts of a consultation (Table 8.3) ⇒ uniform approach. Cards face-down on desk, turned over after greeting & patient's description.
• Patient completed satisfaction questionnaire straight after, and another 1 week later. Consultation time + demographics recorded.
• Check: independent observer analysed 40 random tapes: all aspects covered; intended style identified correctly in 39/40.

• No sig. difference in mean consultation length. Overall high satisfaction – only 3 neutral/negative.
• Directing group more likely to feel the doctor completely understood their problem & the explanation was excellent – straight after and 1 week later (all ratings had dropped at follow-up). Also more likely to feel greatly helped 1 week later.
• After 1 week, for patients who rarely attended, had a physical problem, weren't investigated or got a prescription → directing gave greater feeling of being understood.
• No difference for: consultations > 9 min · main treatment = advice · chronic problems · conditions with a psychological basis.

Conclusion: directing ⇒ higher satisfaction (understanding, explanation, felt improvement) – contradicts the usual view that sharing is better. Best for simple physical illness (traditional biomedical approach); for chronic/psychological problems the advantage disappears ⇒ further research needed.

+ Field experiment – patients thought they were having a normal consult, didn't know it was a study ⇒ high validity, honest answers · closed Qs ⇒ quantitative, comparable · questionnaire immediately and 1 week later ⇒ validity (see change over time) · standardised: same doctor, prompt cards ⇒ only the style differs ⇒ reliable.

– no open Qs ⇒ no qualitative insight · one surgery, 4 months (but both sexes & many ages) · one GP only.

Application: surgeries can advise doctors on style to raise patient confidence/satisfaction – esp. for simple physical conditions.

Instead of one period, break it into three (Fig 8.3):

• Predictors tested: sensory/perceptual experience (pain, bleeding) · self-appraisal (watching for change) · coping response (home remedies) · emotional reactions (fear) · imagined consequences (surgery?) · situational barriers (cost).
• Expected: first three → appraisal delay · emotions & conceptual beliefs + imagined consequences → illness delay · situational & sociodemographic → utilisation delay.
• 93 patients attending one of 4 clinics for first time with a particular illness; wide age/ethnicity, slightly more women. 45-min questionnaire + interview (open Qs + Likert scales).

• No sig. correlation between lengths of the 3 stages. Mean total delay 57 days (skewed by extremes) but median only 8 days.
• 6 variables correlated with total delay:

• Appraisal (avg 4.2 d): pain or bleeding (sensory) ⇒ shorter; reading about the symptom (appraisal strategy) ⇒ longer.
• Illness (avg 3.1 d): a new symptom ⇒ shorter than old/frequent ones (sensory); imagining negative consequences e.g. "on the operating table" (imagery) ⇒ longer.
• Utilisation (avg 2.5 d): worried about cost ⇒ longer (situational); pain ⇒ shorter (sensory); belief symptom can be cured ⇒ shorter (conceptual).

Summary: situational, sensory, conceptual beliefs, sociodemographic & appraisal strategies all correlate with total delay; the stage-by-stage analysis is most useful: appraisal ← sensory + appraisal strategies · illness ← sensory + imagery · utilisation ← sensory + conceptual + situational.

22-year-old female university student, first thought to have immune deficiency.

• Age 17: amenorrhoea (no periods) → medication. Next months: recurring DVT (blood clot in vein) → hospital, anticoagulants. Then painful groin swellings + weak legs ⇒ several investigations, medication change, surgery.
• Latest admission: painful swelling above right breast + several similar abdominal swellings over recent months (≈20 hospital admissions). Abscess drained surgically; days later a lump in the other breast – also drained. Lumps & organisms found couldn't be explained.
• ⇒ Munchausen suspected; psychiatric consult, long-term psychotherapy suggested. She was defensive, extremely rationalising, very stressed. A nurse found evidence she was causing the abscesses herself. Told about another such patient she got angry, left against medical advice, lost to follow-up. Diagnosis confirmed.

• Rates of non-adherence in studies: anywhere from 10% to 92%.
• WHO (2003): ≈ 50% non-adherence to medication in developed countries; ≈ half of that is intentional; more likely for chronic (long-term) than acute (short-term) conditions.
• Missed appointments in primary care: UK 2.9%–11.7%; USA 5%–55%.

Aim: which factors contribute most to rational non-adherence; do medicine/patient background factors matter?

• 161 Australians, online sampling to be representative (age, gender, education, income).
• Survey: (1) current medication use & attitudes (2) Discrete Choice Experiment (DCE): 10 hypothetical situations – imagine taking two medications for different long-term conditions, choose which you'd keep taking (3) background info.
• 8 factors varied: symptom severity · symptom frequency on medication · chance of early death on medication · severity of side effects · chance of future side effects · how medicine is taken · alcohol restrictions · monthly cost.
• 32 choice sets in 4 survey versions + a repeated set (check consistency/rationale).

• 6 of 8 factors sig.; symptom severity & alcohol restrictions did not. Background characteristics and whether currently on medication didn't matter.
• Cost mattered only for those without private insurance (they pay themselves).
• Trade-offs: far more willing to continue if taken once a day than four times – but would accept 4×/day if future side-effect risk fell by 20%.
• 58% weighted harms more than benefits. Risk of current side effects > future side effects. Reducing risk of death > reducing symptom frequency.

Aim: effect of non-participation in adherence studies – expected those who refuse to take part show less adherence.

• Adolescents 11–18 with type 1 diabetes, living with ≥1 family member. 94 families → 52 participants (finished) · 28 non-returners (agreed, didn't return questionnaires) · 14 non-consenters (did interview + demographics, refused rest).
• Measures: adolescent semi-structured adherence interview (→ adherence score) · parent demographic sheet · questionnaires on parent–child relationship (parents ≈ ½ h, teens ≈ 1 h; prepaid return) · medical chart review (blood-sugar tests/day; metabolic control).
• Results: groups didn't differ demographically except father's education (higher in participants). Non-returners did fewer blood-sugar checks/day and scored lower on adherence – still sig. even after blood-sugar items were removed from the score.
• Explanation: adhering & returning forms both need organisation/planning; or non-returners lacked incentive/didn't want to discuss diabetes or family relationships.

+ >1 measure ⇒ consistency/reliability, quantitative. – self-report: lying (hide non-adherence) or memory errors.

Aim: test how well electronic monitoring measures compliance/adherence in asthma. Compliance = passive following of orders (% days correct dose); adherence = collaboration & full understanding.

• 57 asthma patients, 12-week trial, oral zafirlukast: 3 weeks' supply at a time; every 3 weeks a clinic visit to return unused tablets & get a new bottle. 1 tablet morning + 1 evening (≈12 h apart).
• Told compliance was measured – not that the cap was a MEMS. Days classified: insufficient interval (2 events < 8 h apart) · undercompliance (1 event) · no compliance (0) · over-compliance (>2).

• Median adherence 71%; median compliance 89%; 66% of patients had good compliance (>80% of doses). Mean gap between tablets on 2-tablet days: 12 h 34 min.
• Median % of days: one event 19% · insufficient interval 2% · no tablets 1% · over-compliant 2%.
• Tablet count median compliance 92% (slightly above TrackCap). Difference marginal for 83% ⇒ patients occasionally took >1 tablet at a time. A minority removed more than prescribed (one took up to a week's supply per opening); 2 patients showed multiple openings just before clinic visits ⇒ hiding non-compliance from tablet counts.
• Conclusion: compliance high all 12 weeks; TrackCap is an accurate measure.

Methods: small sample (57) but 12 weeks ⇒ valid; knew compliance was measured (ethics) but not the cap's role ⇒ natural behaviour kept; tablet count used to check TrackCap accuracy.

Aim: compare adherence of children with asthma using the Funhaler (spacer + whistle + spinning toy – deep breath makes toy spin & whistle sound ⇒ fun, encourages correct use) vs the standard spacer (holds medicine so it can be breathed in).

• 32 children aged 1.5–6 with asthma (inhaler + standard spacer). 2-week study. Parents did a closed-question questionnaire on attitudes/adherence before and after; ad hoc phone calls asked if child had taken treatment the previous day.

• Fewer problems (screaming, refusal) with Funhaler. Typical response to spacer = fear; to Funhaler no fear. Parents "completely happy": 10% standard vs 60% Funhaler; no parents strongly concerned/disliked it.
• + field experiment ⇒ high ecological validity · repeated measures (same children before/after ⇒ differences not due to participant variables). – no control group (all got Funhaler) · parent self-report ⇒ social desirability · small sample, short time · novelty may wear off ⇒ need bigger, longer study.

• 1970s–80s: methods used to tackle community problems (pollution, energy, dental care, nutrition, seat belts) – not yet immunisations, though uptake was very low in some areas ⇒ children vulnerable, and some schools banned not-fully-immunised children. Later start ⇒ greater risk of not finishing before school age.
• Byrne et al. (1970): computer-generated "more emotional" prompt to half of parents of 2-month-olds (Rhode Island); no reply in 30 days → 2nd prompt → then phone/house call. ⇒ +10% immunisations for lower & upper class, +5% (n.s.) middle class.
• White (1976): radio DJs urged parents; lottery prize (week in Bahamas) – every signed permission slip entered ⇒ 92.6% slips returned.

Aim: evaluate four ways to motivate parents of preschoolers to get children immunised. Prediction (best → worst): specific prompt + money > specific + access > specific alone > general alone.

• Sample: entire population of immunisation-deficient preschool clients (≤5 yrs) at a public health clinic, US city: 2,101 children; 1,133 (53.9%) needed ≥1 immunisation. Even boys/girls, 64% Caucasian, mean 5.2 immunisations needed.

• 3 dependent measures: no. of target children receiving ≥1 immunisation · no. attending clinic for any reason · total immunisations given.
• Effect measured 2 weeks after (when lottery/access applied) + follow-ups at 2 & 3 months.
• Accuracy checks: random 10% of specific prompts – 93% completely accurate · names/addresses 100% · reliability of immunisations recorded by clinic staff 80%.

• Sig. differences between groups on all three measures. Order (high → low): specific + monetary > increased access > specific prompt > general prompt > controls – exactly as predicted; still sig. at 2- & 3-month follow-up. Specific + money ⇒ 29% increase in immunisations given.
• Cost-effectiveness: the specific prompt alone was the most cost-effective per unit; specific + money cost most. Access was least cost-effective at first but became slightly better than money in the long run. Considering size of impact over time, specific prompt + monetary incentive was judged the most cost-effective significant intervention in the long run.
• ⇒ Implications for health departments on how best to run community programmes.

+ large-scale, whole target population ⇒ high population validity · longitudinal (2 wk + 2 & 3 months) shows long-term not just immediate effect · checks on prompt accuracy, mailing addresses & correct immunisations ⇒ validity/reliability · 2 control groups. – self-report/staff-recorded data (80% reliability); one clinic/city, mostly Caucasian ⇒ cultural limits; money may not be appropriate in all cases.

Application: health departments learn general prompts don't work, specific prompts (±incentive) do ⇒ quick, cheap, effective way to raise uptake for a targeted group (and other health campaigns). Individual vs community interventions are cost- & time-effective.

Research-methods hook: could matched pairs replace repeated measures? Note: repeated measures = same children before/after; matched pairs = different but similar participants.

First reported case of mirror treatment for lower-limb PLP. "Alan", 32, life-saving surgery removed his leg at the hip; very unwell for weeks, fully aware of amputation only 5 weeks later.

• Within 2 days: PLP + pain at amputation site. Mild at start of day ('pins and needles', crossed toes) → severe by late afternoon. Phantom felt shorter than real leg, "in a cast", stretching backwards with toes pointing down. Painkillers & TENS had little effect ⇒ tried mirror.
• Procedure: 10 exercises × 10 reps (straighten/bend leg, foot up/down, clench/unclench toes, foot circles). Twice a day with physio at first → alone → eventually without mirror (up to 4×/day). 3 weeks.
• Some exercises harder; took 4 sessions before he felt any control; emotional seeing reflected leg.
• Ratings: phantom pain 5–9 → 0 · stump pain 0–2 → 1 · sense of control of phantom leg 0–3% → 25–30%; crossed-toes sensation gone by week 3; phantom still shorter than real leg but could be "straightened out" with effort.

Conclusion: mirror treatment works for lower limbs as for upper.

Assesses quality & intensity (cancer, muscular pain…). 78 words; patient picks those that fit; words weighted by severity ⇒ total 0 (no pain) – 78 (severe).

• Descriptors: flickering, sharp, searing, tender, tugging · Affective: tiring, sickening, terrifying, wretched · Evaluation: annoying, troublesome, intense, unbearable · Misc: spreading, numb, squeezing.
• Also: what increases/decreases pain (eating, heat, cold, weather, movement, rest, mild exercise) + strength now / worst / least: mild · discomforting · distressing · horrible · excruciating.

+ tracks change over time/treatment effectiveness · quantitative ⇒ stats & comparison · quick & easy. – closed Qs may force an answer that doesn't fit ⇒ no qualitative detail.

• Grant (2006): children's pain often under-managed, esp. in outpatient/emergency settings – analgesia used too rarely, delayed, dose too low. Reasons: time pressure, fear of reduced productivity, reluctance (side effects) (Fein 2012; Brown 2003). Poor pain management ⇒ slower healing, later anxiety/fear of medical care.
• Singer et al. (2009): parents' & practitioners' ratings are generally lower than the child's. Morrow et al. (2012): chronic illness – children's quality-of-life reports differ from doctors' & parents'.

Aims: ① agreement on pain intensity between child, parent, physician ② influence of child's age, condition, severity on the differences ③ how assessments affected physicians' administration of pain relief.

• 243 children aged 3–15 (53% boys), Norwegian emergency department, 17 days. Child, parent & treating physician each completed a questionnaire. 51 physicians (57% men; 51% had own children; half > 5 yrs experience; 30% family-medicine speciality; none paediatric).
• Age-appropriate scales: 3–8 yrs: FPS-R (6 faces) · 9–15 yrs: VAS and Coloured Analogue Scale (green no pain → red worst) · parents & physicians: Numeric Rating Scale (0–10).
• Parents completed theirs before the child but weren't fully blind; told not to tell physician. Parents' form = demographics; physicians' = experience, speciality, own children. Diagnosis types: infection, fracture, wound, soft-tissue/ligament/muscle (51% soft tissue).

• Mean NRS: physicians 3.2 · parents 4.8 · children 5.5. Physicians' scores were 2.3 lower than the child's, 1.6 lower than parents'.
• Agreement: parents–physicians 15% · children–physicians 14.6% · parents–children 40.1%.
• Parents & doctors rated pain lower for younger children (bigger gap between child and adult for younger children).
• By severity (physician – child difference, NRS): mild 3.2 · moderate 1.2 · severe –0.6 (physician sometimes rated higher than child). When physician felt child's reaction fitted the condition (concordance) gap smaller (2.0 vs 3.1).
• No link to waiting time, child's gender, physician's gender/experience, having own children.

• Painkillers given to only 8.6% of children (mean NRS 4.9 by physician). By physician's rating: mild 4.1% · moderate 9% · severe 42%. By the child's own rating only 14.3% of children with severe pain got painkillers.
• Conclusion: emergency physicians significantly underestimate children's pain vs parents & children – all conditions & ages (less so for ≥ 8 yrs). Parents' assessments are much closer to the child's (though still lower).

+ relatively large sample, wide age/condition range, many physicians ⇒ validity · real-life ED ⇒ ecological validity. – self-report (doctors may adjust ratings because they know they're in a study).

Application: educate doctors about under-rating ⇒ more accurate assessment ⇒ more painkillers ⇒ less pain and less later fear of hospitals; doctors can rely on parents' judgement when child can't give a rating.

10 target behaviours; an observer who lives with the patient records each daily (severity/frequency). Each = 0, ½ or 1 ⇒ total /10 (high = more pain behaviour & impairment).

+ info on outward signs others can observe. – relies on observer interpretation · observer may miss behaviours · scores don't correlate well with the McGill ⇒ displayed pain ≠ felt pain (conditioned/reinforced pain behaviour; or someone hides pain to cope/protect others).

Very fine needles (4–10) in specific skin points, left 10–30 min, several sessions. Headaches, back & nerve pain. Believed to release endorphins (body's natural painkillers) and ↑serotonin. Evidence mixed – maybe placebo; but side effects extremely rare ⇒ little harm.

Small device, electrodes on painful area deliver a mild current (tingling). ↓pain signals to brain & spinal cord, relaxes muscles, may ↑endorphins. Arthritis, sports injuries, childbirth. Reliability not firmly shown; works well for some people/conditions; only temporary relief while in use; very safe.

SRRS given to 2,500 male US sailors for life events in the previous 6 months; their health recorded during the following 6-month tour of duty. Significant but small positive correlation between life-change score and illness score (more life change ⇒ more illness).

• 10,000+ male & female London civil servants, 1985–2004; 7 phases: postal questionnaires, alternate phases with clinical exam.
• Job strain = high demands + low control. Iso-strain = job strain + social isolation at work (no supportive co-workers/supervisors). Total work-stress score = no. of strain/iso-strain reports in phases 1 & 2.
• Results: ages 37–49: sig. link between work stress & CHD events; little link at 50+. Younger: work stress ⇒ poorer health behaviours (less fruit/veg, more alcohol, less activity). Adjusting for behaviours only slightly reduced the link ⇒ ≈32% of effect explained by unhealthy behaviour; rest = physiological (↑cortisol, ↑BP).

3,000 men aged 39–59; questionnaire ⇒ Type A or B (≈ half each); followed 8½ years. Result: more than twice as many Type A as Type B developed CHD. Of 257 who developed CHD, 70% were Type A. Adjusted for smoking etc. ⇒ Type A still almost twice as likely.

Aim: use fMRI to measure CBF changes to mild–moderate stress from a maths task.

• 30 Americans: 23 stress group (mean age 24, 11 F), 7 control (mean 23, 4 F). All tested 3–5 pm (control for daily cortisol variation).
• 4 scans × 8 min: baseline → low-stress task (count aloud backwards from 1000 – controls auditory/verbal activation) → high-stress task (subtract 13 from a 4-digit number, prompts to hurry, restart on error) → baseline.
• Measures: self-reported stress & anxiety (1–9), saliva cortisol, effort/frustration/difficulty (1–9), heart rate (pulse oximeter). Control group: no tasks, same HR/saliva/self-report (checks scanner itself isn't stressful).

• Self-reported stress/anxiety, HR & cortisol all ↑ in the high-stress task, ↓ in second baseline (cortisol lags ≈10 min).
• Right prefrontal cortex (RPFC) – linked to negative emotion (sadness, fear) & vigilance to threat. Positive correlation (r = 0.65) between change in RPFC CBF and subjective stress rating; RPFC CBF higher with higher cortisol & anxiety ⇒ brain activity matches both subjective & objective stress.
• RPFC activity stayed high after the task ⇒ stress ⇒ prolonged vigilance. Difficulty & effort were not linked to RPFC change ⇒ it's stress specifically, not just mental effort.

• 139 adult commuters (New Jersey → Manhattan). Saliva at end of morning trip and at home on following weekend. Also: motivation (persistence in proofreading task – errors found), mood (two 5-point scales: carefree–burdened, contented–frustrated).
• Crowding: distal (density of whole carriage) vs proximal (people on the same row).
• Results: distal crowding – no sig. effect; proximal crowding sig. ↑ stress on all three (cortisol, proofreading, mood) ⇒ only crowding in the area immediately around your seat matters.

Forehead = hard muscle to relax deeply ⇒ if biofeedback works here, likely works elsewhere. 15 participants lay quietly, eyes closed, relaxing forehead; 20 trials 1–3 days apart, same time each day; randomly assigned to:

• Feedback: tone changes with muscle tension (low tone = relaxed) – know they're doing right.
• Irrelevant feedback: constant low tone (told it will help).
• No feedback: silence; told to relax.

• Relaxation (McLean 1986): focus on muscle relaxation + deep breathing ⇒ slower breathing, calmer, ↓tension ⇒ ↓stress. Tighten a muscle group, hold, notice tension, then relax slowly and notice the relaxed feeling. Deep slow breathing (too fast ⇒ dizzy & anxious): slow breath in, slow breath out.
• Imagery (Lyles et al. 1982): imagine a peaceful scene (real/imaginary; safe, calm, happy) in as much sensory detail as possible, eyes closed, combined with relaxation & breathing ⇒ calm, ↓stress.

• Cancer diagnosis + treatment + uncertain future = stress (fear of losing "health, role and life" – Maguire 1981). Reducing stress ⇒ more positive outlook & more emotional resources to cope. Relaxation imagery helped hypertension & cancer patients (Simonton & Simonton 1975).

Aims / hypotheses: relaxation & imagery would both reduce stress vs no treatment; women given relaxation + imagery would benefit more than relaxation alone.

• 139 women, outpatients on a 6-week radiotherapy course after early breast cancer; all under 70; one London hospital. Told they'd be randomly assigned to one of 3 conditions & fill in mood questionnaires at start (after 1 radiotherapy session) and after 6 weeks.
• Conditions (random allocation): ① control (no treatment) ② relaxation ③ relaxation + imagery.
• Treatment groups taught relaxation (directed sensory awareness on muscle groups + deep breathing); group ③ also taught to imagine a peaceful scene. A tape to play 15 min every day repeating instructions. All groups saw researcher ½ hour once a week – treatment groups practised, controls talked about themselves & interests.
• Measures: Profile of Mood States (65 items: tension, fatigue, anger, confusion ⇒ total mood disturbance) and Leeds General Scale (6 anxiety + 6 depression items, 4-point scale; for people not formally diagnosed).

• Change in total mood disturbance differed sig. between groups (not the Leeds scale).
• Mood improved in relaxation and relaxation + imagery groups – greatest with relaxation + imagery; the control group's mood actually worsened.
• Both treatments more effective for women aged 55+ (esp. ↓depression & tension scores). Women high in anger at the start responded less (maybe anger blocked benefit).

Conclusion: relaxation, and particularly relaxation + imagery, have a positive effect on mood state.

+ reasonable sample · all same hospital/stage ⇒ illness/stage not a confounding variable · random allocation ⇒ ↓participant bias · control group ⇒ valid comparison, pre- & post-measures · closed Likert questionnaires: quick (women already doing radiotherapy), quantitative for stats.

– all women, one cancer, one hospital ⇒ not generalisable to men/other conditions · no qualitative data (no detail of feelings) · self-report.

Application: advise other cancer patients that relaxation may help; less stress ⇒ feel better and more mental energy for treatment (could work for other conditions/men too). Ethics (cost–benefit): controls got no stress management and their mood declined, while treatment groups improved – do costs to controls outweigh benefits of findings?

Think: how could idiographic vs nomothetic apply? (individual case detail vs large-group stats).

Aim: effects of fear appeals in impersonal group communication + possible adverse (defensive) reactions.

• 200 high-school students; recorded lecture + slides on dental hygiene – same length/speaker/info on tooth decay causes & oral-hygiene advice; only fear level differed:
• Form 1 strong: painful consequences of decay/gum disease; personalised ("this could happen to you"); 11 very graphic photos. Form 2 moderate: milder, factual, impersonal; 9 milder photos. Form 3 minimal: hardly any negative consequences – talk about tooth growth/function; only X-rays & diagrams. + control (lecture on different topic). Groups matched on age, sex, education, IQ.
• Questionnaires: 1 week before (inside wider health Qs), immediately after, and 1 week later.

• Emotion: "worried a few/many times about own mouth": 74% strong · 60% moderate · 48% minimal. Concern scores: 7.8 / 6.6 / 5.9 (strong sig. > others; moderate > minimal).
• Behaviour (5 teeth-cleaning Qs, 2 wks after): minimal group improved most; strong group least – the opposite of "fear motivates". Dentist visits next week: minimal 18% · moderate 14% · strong 10%. (only shows they say they did it – not proof)
• Resistance to contradictory propaganda (paragraph saying toothbrush type doesn't matter): all 3 groups showed some resistance vs control; minimal group significantly more resistant.

Conclusion: minimal appeal = most effective – ↑resistance to counter-propaganda & ↑self-reported adherence. Strong fear doesn't ↑acceptance of advice – seems to have the opposite effect (defensive avoidance).

+ control group + 3 matched groups ⇒ changes due to the manipulation. – only self-report of behaviour. Think (Q13): strong fear is upsetting ⇒ people avoid/deny the message.

• 190 patients admitted after a confirmed myocardial infarction (MI – heart attack: blood supply to heart suddenly blocked). 3 days after MI assigned to control or experimental, stratified (age, social class, initial distress); person holding the list blind to purpose; treating staff blind to group.
• Experimental: on discharge got "the manual" – 6 weekly sections: education, home exercise programme, recorded relaxation/stress management, self-help for intrusive/distressing thoughts, anxiety, depression. Introduced by a facilitator (+ partner). Control: time with facilitator + leaflets on recovery.
• Contact at 1, 3, 6 weeks (brief call/clinic): experimental – progress check & encouragement to stick to exercises; control – general encouragement, questions answered. Anxiety/depression/general psychological health by questionnaire at 6 weeks, 6 months, 1 year; GP surgeries reported contacts at 6 months & 1 year.

• Anxiety dropped sig. more in experimental group (6 weeks & 1 year). Depression differed sig. only at 6 weeks. "Caseness" (poor psychological health) sig. lower in experimental at 6 wks, 6 mths & 1 yr.
• For especially distressed patients at time of MI: control group had higher anxiety & depression at every follow-up.
• Doctor visits & hospital readmissions higher in controls – both first 6 months and following 6 months.

Conclusion: information + a self-help programme significantly improves recovery after MI and psychological health.

+ high ecological validity (real patients in real recovery); control group; blind staff; stratified groups ⇒ equal. – only MI patients (but similar programmes could be designed for other conditions).

• Home study: 5–6-year-old fussy eaters who rarely ate fruit/veg. 4 procedures (fruit then veg): presentation only · rewarded taste exposure · peer modelling · rewarded exposure + peer modelling. Peer modelling = video of slightly older kids ("Food Dudes") get superpowers from fruit/veg and fight "Junk Punks"; rewards = Food Dude stickers/pens for target amounts.
• Start: children ate avg 4% of fruit & 1% of veg offered. Modelling + rewards ⇒ 100% fruit, 83% veg; 6 months later (no rewards/modelling) still 100% & 83%, and trying other fruit/veg. Presentation alone: no effect; modelling without rewards: minimal; rewards only: some effect (esp. fruit) but far below combined.

• Nursery (ages 2–4): video modelling with 2 cartoon characters + rewards at mid-morning snack (fruit then veg); then maintenance with intermittent rewards. Fruit 30% → 71%; 15-month follow-up 79% (lunchtime fruit 17% → 79%). Veg 34% → 87%, still 86% at 9 months; 3 years later ≈ 80% of fruit & veg eaten.
• Schools (fruit/veg offered with crisps & chocolate): fruit more than doubled 28% → 59%; veg 8% → 32% (6-month follow-up). Home-based subset similar.
• "Whole school" programme (ages 4–11): Food Dude video, rewards, home pack, letters, staff info; 16-day intervention then maintenance (no videos; letters/rewards intermittent). 3 schools: large sig. increases across all schools, ages, boys & girls; parents reported ↑ at home. Commissioned by the UK FSA & DoH for hundreds of schools.

Why it works (3 reasons): ① repeated tasting ⇒ real liking (continues without rewards) ② school culture – eating fruit/veg socially reinforced by peers ③ self-concept – "I'm a fruit & veg eater" guides behaviour.

Mining = dangerous: hundreds of deaths, thousands of injuries/year in USA; lost time = lost money. A token economy = reward system: tokens for specific desirable behaviour, exchanged for goods.

• 2 open-pit mines (similar methods, heavy machinery, high injury rates in prior 5 yrs). Employees in all areas incl. management/clerical (≈650 up to 11,000 over the years – as printed). Standard H&S (training, refreshers, inspections) already in place.
• Workers in 4 groups by previous lost-time injuries: G1 least hazardous (≥75% office) · G2 foremen, supervisors, technicians, surveyors · G3 mechanics, labourers, machine operators · G4 most hazardous (electricians, scraper operators, fuel workers).
• Token economy: trading stamps given each month with pay for every month without a lost-time or compensation injury; most stamps for G4, fewest for G1. Extra stamps if everyone under a supervisor injury-free; stamps for adopted safety suggestions & acts preventing serious injury/equipment damage. Injured employee lost stamps for 1–6 months (depending on days missed); also for damage caused.
• Stamps = secondary reinforcers (no value alone) swapped at redemption stores/catalogue for primary reinforcers (items with intrinsic value: spice racks → microwaves → gas barbecues).

• Measured: frequency rate (injuries causing ≥1 day absence), severity rate (hours missed), costs of damage/compensation/medical care; cost–benefit analysis (money saved minus costs).
• Days lost → 11% and 2% of baseline (the two mines). Lost-time injuries → 15%–32% of baseline. Costs of accidents ↓ ≈ 90%.
• Saving: one mine $18.1–27.8 per $1 spent; other $12.9–20.7. Last 10 yrs: days lost = ¼ of national average at one mine, 1/12 at the other. Employees also benefited from stamps.

Behavioural programmes are effective & maintained for long periods. Think (Q14): workers got goods & safer jobs; owners saved money (fewer lost days, lower costs). Activity 8.8 (building site): tokens for accident-free months, rewards for safety suggestions, group bonuses, redemption catalogue.

Aim: test that people think negative events are less likely for them, positive more likely, plus factors: intensity, perceived probability, personal experience, controllability, stereotype.

• Diverse college students judged 18 positive (own home, living past 80, marrying someone wealthy, gifted child) + 24 negative (fired, heart attack < 40, car injury, sued) events compared with same-sex students at their college: from "100% less (no chance)" … through "3× / 5× average". Other groups rated % who'd experience it, controllability, desirability, personal experience, and stereotype salience.
• Results: average positive events +15.4% above zero; negative –20.4% ⇒ unrealistic optimism. Comparative judgements higher for: more desirable · higher perceived probability · prior personal experience; lower for negative events with high controllability or a stereotype attached; higher for positive with high controllability.

Aim: test whether unrealistic optimism arises from an inaccurate/incomplete image of others; also check Study 1 wasn't confusion between "overall likelihood" and "likelihood for me".

• 120 females (from Study 1 raters). Everyone made comparative judgements for 10 events (pre-treatment), then 3 random conditions: experimental – listed factors that affect own chance, then got 5 lists made by others, then judged the remaining events · control 1 – listed own reasons, no others' lists · control 2 – no lists until after judging. Half in each group got a cover sheet warning not to confuse comparative with overall probability.
• Results: warning had no effect (not confused). Experimental group sig. less optimistic on negative events than both controls. For positive events experimental & control 1 both less optimistic than control 2 ⇒ just listing own factors lowers optimism (first thoughts optimistic; later reflection realistic). Optimism not eliminated – still unrealistic. More favourable than unfavourable reasons ⇒ more optimism (all groups).

Unrealistic optimism comes from an inaccurate view of other people.

Methods: self-report ⇒ demand characteristics (students may guess aim, change answers), not wanting to look silly, misunderstanding. + Study 2 tackled confusion with clear instructions (no difference ⇒ not a problem) ⇒ ↑validity. – all college students from one place; younger people may differ from older adults. Think (Q15): to check Study 1 results weren't due to confusion & to test why optimism occurs.

Aim: can a science of positive psychology lead to happiness? Annual seminar for UPenn students.

• Start: detailed personal introductions – listening skills to find positive strengths in classmates. Next 4 sessions: documented science of positive emotion – past (contentment, satisfaction, serenity), future (optimism, hope, trust, faith), present (joy, comfort, pleasure). Topics: depressive realism (happy people may be less accurate) · wealth & life satisfaction (richest no happier than average) · set ranges (lottery winners & paraplegics return to baseline within a year; capacity for pleasure ≈ 50% inherited). Techniques: dispute unrealistic catastrophic thoughts ⇒ ↑optimism; practise gratitude.
• Pleasant life (present): savouring (sharing experiences, mementos) + mindfulness (new angles, slow down).
• Good life: pleasure = felt, conscious, fades fast (food, hot shower) vs gratification = longer-lasting, absorbing, retrospective "wow!" (volunteering, rock-climbing). Use VIA questionnaire to find top 5 strengths (kindness, perseverance, love of learning, spirituality…) & use them daily (homework).
• Meaningful life: use highest strengths to serve something larger (family, college, workplace); family trees of strengths; imagine a positive human future; read Aging Well (Vaillant) – 60-year study: Harvard graduates no better than lower-status men on life satisfaction, marital happiness, maturity, freedom from depression/alcoholism, longevity.

Seligman: thousands spent on education may lead to less life success than learning this material – "we learn more when lighting candles than cursing the darkness".

• Costello et al. (2004): dramatic rise in child/adolescent mental health disorders since late 1990s. WHO (2005): rates under-18s 8–20%. Huebner et al. (2000): 25% of US students reported an "unhappy/terrible" existence. School culture shifted to academic achievement over social/emotional aspects (Hargreaves 2003).
• Dual-factor model (Greenspoon & Saklofske 2001): mental illness and positive mental health are two distinct dimensions – you can have high distress and high well-being. Subjective well-being = cognitive (life satisfaction) + affective (positive feeling, absence of negative emotion); linked to achievement, engagement, relationships (Antaramian 2010; Gilman & Huebner 2006).
• Hedonic approach (pleasure, life satisfaction, happiness) vs eudaimonic (realising potential, meaning, autonomy, purpose) (Ryan & Deci 2000).

Aims: will the intervention ↑ self-efficacy, optimism, life satisfaction, self-esteem and ↓ distress/mental-health problems? Demographic differences (gender, income, single vs two-parent)?

• 1,167 students aged 11–14 in two large middle schools, centre of Israel (35 schools approached; 8 principals interested; one randomly → intervention, a demographically matched school = control – got the programme after). Almost all Jewish (varied observance), mixed SES, 64% two-parent / 36% single-parent.
• Measures: Brief Symptom Inventory (BSI) – 53 self-report items (4-point Likert), 10 subscales (depression, anxiety, paranoid ideation, hostility, interpersonal sensitivity…) · Rosenberg self-esteem scale · general self-efficacy · SWLS life satisfaction · LOT-R optimism.
• Intervention (2 strands, whole school year): teachers attended 15 two-hour workshops by clinical psychologists (group dynamics + positive psychology); teachers ran an age-appropriate class programme with textbook & materials – activities, discussions, video clips, poems & stories. E.g. gratitude: list 5+ things grateful for that week & share. Control: usual lessons.
• Questionnaires: Sept 2010 (before), June 2011 (end), follow-ups Dec 2011 & June 2012.

• At start: lower income / single-parent ⇒ more anxiety/depression, lower self-esteem & life satisfaction. Boys: more anxiety, lower self-efficacy, optimism & life satisfaction; girls: more other poor-mental-health signs, lower self-esteem.
• Intervention vs control: intervention ⇒ fewer mental health symptoms over time (control symptoms rose): sig. ↓ general distress, depression, anxiety, interpersonal sensitivity; sig. ↑ self-esteem, self-efficacy, optimism (control: ↓ self-esteem & self-efficacy, only slight ↑ optimism). Unexpected: no sig. difference in life satisfaction.
• High-risk students (low income / single-parent) improved less than low-risk, but still improved.

Conclusion: a successful application of positive psychology; improves mental health symptoms; also positive effect on whole-school culture (staff–student relationships).

+ matched control school ⇒ valid comparison · longitudinal (see change in both groups over time) · school setting, run by staff ⇒ ecological validity, repeatable for future year groups. – intervention in one school ⇒ caution generalising · self-report only ⇒ social desirability & demand characteristics (students know the "right" answer after the programme). How else collect data? – teacher/parent ratings, observation, interviews, school records.

Application: other schools can run it; adolescent mental health worry is rising ⇒ helps students and staff, effects for years.